{"id":74018,"date":"2025-02-07T16:47:50","date_gmt":"2025-02-07T16:47:50","guid":{"rendered":"https:\/\/peraltafinancing.com\/analytics\/andrew-ngs-visionagent-streamlining-vision-ai-solutions\/"},"modified":"2025-02-07T16:47:50","modified_gmt":"2025-02-07T16:47:50","slug":"andrew-ngs-visionagent-streamlining-vision-ai-solutions","status":"publish","type":"post","link":"https:\/\/fivemor.com\/?p=74018","title":{"rendered":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions"},"content":{"rendered":"<p> <br \/>\n<\/p>\n<div id=\"\">\n<table class=\"table table-bordered border-black table-striped\">\n<tr>\n<td><strong>Model<\/strong><\/td>\n<td><strong>Recall<\/strong><\/td>\n<td><strong>Precision<\/strong><\/td>\n<td><strong>F1 Score<\/strong><\/td>\n<\/tr>\n<tr>\n<td><strong>Landing AI<\/strong><\/td>\n<td>77.0%<\/td>\n<td>82.6%<\/td>\n<td><strong>79.7%<\/strong> (highest)<\/td>\n<\/tr>\n<tr>\n<td><strong>Microsoft Florence-2<\/strong><\/td>\n<td>43.4%<\/td>\n<td>36.6%<\/td>\n<td>39.7%<\/td>\n<\/tr>\n<tr>\n<td><strong>Google OWLv2<\/strong><\/td>\n<td>81.0%<\/td>\n<td>29.5%<\/td>\n<td>43.2%<\/td>\n<\/tr>\n<tr>\n<td><strong>Alibaba Qwen2.5-VL-7B-Instruct<\/strong><\/td>\n<td>26.0%<\/td>\n<td>54.0%<\/td>\n<td>35.1%<\/td>\n<\/tr>\n<\/table>\n<h3 class=\"wp-block-heading\" id=\"h-4-key-takeaways\">4. Key Takeaways<\/h3>\n<ul class=\"wp-block-list\">\n<li><strong>Landing AI\u2019s Agentic Object Detection<\/strong> achieved the highest F1 Score (79.7%), meaning it balances precision and recall better than the others.<\/li>\n<li><strong>Google OWLv2<\/strong> had the highest recall (81.0%), meaning it detects the most objects but has lower precision.<\/li>\n<li><strong>Alibaba\u2019s Qwen2.5-VL-7B-Instruct<\/strong> had relatively low recall but moderate precision.<\/li>\n<li><strong>Microsoft Florence-2<\/strong> showed lower performance across all metrics compared to the other models.<\/li>\n<\/ul>\n<p>Landing AI\u2019s approach seems to provide the most balanced and effective performance in object detection tasks, as reflected in its highest F1 score. The other models have trade-offs, with some excelling in recall but lacking in precision, or vice versa.<\/p>\n<h2 class=\"wp-block-heading\" id=\"h-practical-working-of-visionagent\">Practical Working of VisionAgent<\/h2>\n<p>VisionAgent follows a structured process to provide outputs effectively:<\/p>\n<ul class=\"wp-block-list\">\n<li>Upload an image you want to analyze<\/li>\n<li>Write a prompt (e.g., \u201cdetect people with glasses\u201d)<\/li>\n<li>Our AI agent analyzes the image thoroughly<\/li>\n<li>Receive detection results on your image<\/li>\n<\/ul>\n<h2 class=\"wp-block-heading\" id=\"h-1-prompt-detect-vegetables-in-and-around-the-basket\">1. Prompt: \u201cDetect vegetables in and around the basket\u201d<\/h2>\n<h3 class=\"wp-block-heading\" id=\"h-step-1-conversation\">Step 1: Conversation<\/h3>\n<p>A user initiates a request, describing the vision task in natural language. VisionAgent interprets this prompt and confirms the task.<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-input-image\">Input Image<\/h4>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"612\" height=\"408\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-6.webp\" alt=\"Input Image\" class=\"wp-image-220345\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-6.webp 612w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-6-300x200.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-6-150x100.webp 150w\" sizes=\"auto, (max-width: 612px) 100vw, 612px\"\/><\/figure>\n<\/div>\n<h4 class=\"wp-block-heading\" id=\"h-conversation\">Conversation<\/h4>\n<p>\u201cI\u2019ll help you write code to detect vegetables in and around the basket in the image. I\u2019ll use object detection to identify various vegetables in the scene.\u201d<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-step-2-planner\">Step 2: Planner<\/h3>\n<p>I need to first understand what\u2019s in the image and get suggestions on how to detect vegetables in and around the basket. I\u2019ll start by loading the image and using claude35_vqa and suggestion functions.<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-planning-steps\">Planning Steps<\/h4>\n<ul class=\"wp-block-list\">\n<li>Understand the image contents using <strong>VQA (Visual Question Answering)<\/strong>.<\/li>\n<li>Generate relevant suggestions for the best detection approach.<\/li>\n<li>Select tools that suit the task, such as object detection and color-based classification.<\/li>\n<\/ul>\n<h4 class=\"wp-block-heading\" id=\"h-example-plan\">Example Plan<\/h4>\n<ul class=\"wp-block-list\">\n<li>Identify all objects in the image.<\/li>\n<li>Determine which objects are vegetables.<\/li>\n<li>Differentiate vegetables inside and outside the basket.<\/li>\n<li>Use a <strong>nested detection structure<\/strong> to analyze the spatial relationships.<\/li>\n<\/ul>\n<h4 class=\"wp-block-heading\" id=\"h-execution\">Execution:<\/h4>\n<pre class=\"wp-block-code\"><code>image = load_image('vegetablebasket.jpg')\nclaude35_vqa('Can you describe this image? What vegetables do you see in and around the basket?', [image])\nsuggestion('How can I detect vegetables in and around the basket in this image?', [image])<\/code><\/pre>\n<h4 class=\"wp-block-heading\" id=\"h-observation\">Observation<\/h4>\n<pre class=\"wp-block-code\"><code>[claude35_vqa output]\nThis image shows a beautiful array of fresh vegetables arranged around and in a wicker basket. The vegetables include:\n\n1. Fresh carrots with green tops\n2. Cauliflower\n3. Onions\n4. Potatoes\n5. Broccoli\n6. Eggplant\n7. Red tomatoes\n8. Yellow squash\/zucchini\n9. Purple cabbage\n10. Green bell pepper\n11. Red bell pepper\n12. Cucumbers\n13. Fresh garlic\n14. Fresh herbs (appears to be parsley)\n15. Green banana peppers or chili peppers\n\nThe vegetables appear to be very fresh and vibrant in color, arranged on what looks like a wooden surface or cutting board. The composition shows a nice variety of colors, from deep purples to bright reds, oranges, and greens, making it visually appealing and suggesting a healthy, nutritious selection of produce.\n[end of claude35_vqa output]\n\n[suggestions]\n\n1. The vegetables are both in and around a basket, so relative_position would help in determining their locations with respect to the basket as an anchor point.\n\n2. nested_structure applies because some vegetables are nested\/contained within the basket while others are arranged outside it.\n\n3. color would be useful since the vegetables are distinctly colored (red tomatoes, orange carrots, green broccoli, purple cabbage, etc.), making color a key identifier.\n\n4. The problem does not involve small text, grid patterns, tracking in video, comparing sizes specifically, or dealing with extremely small objects in a large image. The depth position is visible but not crucial to the detection task.\n\n\n[suggestion 0]\nYou need to find the color of objects in the image, you can use the following code to help with this task:\n\nimport numpy as np\nimport cv2\n\ncolor_ranges = {\n    \"red_lower\": ((0, 100, 100), (int(179 * 20 \/ 360), 255, 255)),\n    \"orange\": ((int(179 * 21 \/ 360), 100, 100), (int(179 * 50 \/ 360), 255, 255)),\n    \"yellow\": ((int(179 * 51 \/ 360), 100, 100), (int(179 * 70 \/ 360), 255, 255)),\n    \"green\": ((int(179 * 71 \/ 360), 100, 100), (int(179 * 150 \/ 360), 255, 255)),\n    \"cyan\": ((int(179 * 151 \/ 360), 100, 100), (int(179 * 180 \/ 360), 255, 255)),\n    \"blue\": ((int(179 * 181 \/ 360), 100, 100), (int(179 * 265 \/ 360), 255, 255)),\n    \"purple\": ((int(179 * 266 \/ 360), 100, 100), (int(179 * 290 \/ 360), 255, 255)),\n    \"pink\": ((int(179 * 291 \/ 360), 100, 100), (int(179 * 330 \/ 360), 255, 255)),\n    \"red_upper\": ((int(179 * 331 \/ 360), 100, 100), (179, 255, 255)),\n    \"white\": ((0, 0, 200), (179, 25, 255)),\n    \"gray\": ((0, 0, 50), (179, 50, 200)),\n    \"black\": ((0, 0, 0), (179, 255, 30)),\n}\n\ndef get_color(image, color_ranges):\n    hsv_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)\n    detected_colors = {}\n    for color, (lower, upper) in color_ranges.items():\n        upper_range = np.array(upper, dtype=np.uint8)\n        lower_range = np.array(lower, dtype=np.uint8)\n        mask = cv2.inRange(hsv_image, lower_range, upper_range)\n        detected_pixels = cv2.countNonZero(mask)\n        detected_colors[color] = detected_pixels\n\n    if \"red_lower\" in detected_colors and \"red_upper\" in detected_colors:\n        detected_colors[\"red\"] = detected_colors[\"red_lower\"] + detected_colors[\"red_upper\"]\n        del detected_colors[\"red_lower\"]\n        del detected_colors[\"red_upper\"]\n    return sorted(detected_colors, key=detected_colors.get, reverse=True)[0]\n\n[end of suggestion 0]\n[suggestion 1]\nYou are trying to count objects within objects, or a nested structure. You can solve this by first detecting the outer objects, then cropping the image to the bounding box of each outer object and detecting the inner objects. You can use the following code to help with this task:\n\nall_dets = <an object=\"\" detection=\"\" tool=\"\">(\"object\", image)\n\nheight, width = image.shape[:2]\n\ndef area(box):\n    return (box[2] - box[0]) * (box[3] - box[1])\n\n# only check inner detections on top 25 largest outer detections\nlargest_dets = sorted(dets, key=lambda x: area(x[\"bbox\"]), reverse=True)[:25]\nfor det in largest_dets:\n    x1 = int(det[\"bbox\"][0] * width)\n    y1 = int(det[\"bbox\"][1] * height)\n    x2 = int(det[\"bbox\"][2] * width)\n    y2 = int(det[\"bbox\"][3] * height)\n\n    crop = image[y1:y2, x1:x2]\n    crop_height, crop_width = crop.shape[:2]\n\n    inner_dets = <an object=\"\" detection=\"\" tool=\"\">(\"object\", crop)\n    for inner_det in inner_dets:\n        x1_inner = int(inner_det[\"bbox\"][0] * crop_width)\n        y1_inner = int(inner_det[\"bbox\"][1] * crop_height)\n        x2_inner = int(inner_det[\"bbox\"][2] * crop_width)\n        y2_inner = int(inner_det[\"bbox\"][3] * crop_height)\n\n        bbox = [\n            x1 + x1_inner,\n            y1 + y1_inner,\n            x1 + x2_inner,\n            y1 + y2_inner,\n        ]\n        norm_bbox = [\n            bbox[0] \/ width,\n            bbox[1] \/ height,\n            bbox[2] \/ width,\n            bbox[3] \/ height,\n        ]\n        all_dets.append(\n            {\n                \"label\": inner_det[\"label\"],\n                \"score\": inner_det[\"score\"],\n                \"bbox\": norm_bbox,\n            }\n        )\n\n[end of suggestion 1]\n[suggestion 2]\nYou are trying to locate an objects relative to 'anchor' objects. The 'anchor' objects can be detected fine, but there are many of the other objects and you only want to return the ones that are located relative to the 'anchor' objects as specified by the user. You can use the following code to help with this task:\n\n# First find a model that can detect the location of the anchor objects\nanchor_dets = <a model=\"\" that=\"\" can=\"\" detect=\"\" the=\"\" location=\"\" of=\"\" anchor=\"\" objects=\"\">(\"anchor object\", image)\n# Then find a model that can detect the location of the relative objects\nrelative_dets = <\/a><a model=\"\" that=\"\" can=\"\" detect=\"\" the=\"\" location=\"\" of=\"\" relative=\"\" objects=\"\">(\"relative object\", image)\n\n# This will give you relative objects 'above' the anchor objects since it's the\n# distance between the lower left corner of the relative object and the upper left\n# corner of the anchor object. The remaining functions can be used to get the other\n# relative positions.\ndef above_distance(box1, box2):\n    return (box1[\"bbox\"][0] - box2[\"bbox\"][0]) ** 2 + (\n        box1[\"bbox\"][3] - box2[\"bbox\"][1]\n    ) ** 2\n\ndef below_distance(box1, box2):\n    return (box1[\"bbox\"][0] - box2[\"bbox\"][0]) ** 2 + (\n        box1[\"bbox\"][1] - box2[\"bbox\"][3]\n    ) ** 2\n\ndef right_distance(box1, box2):\n    return (box1[\"bbox\"][0] - box2[\"bbox\"][2]) ** 2 + (\n        box1[\"bbox\"][1] - box2[\"bbox\"][1]\n    ) ** 2\n\ndef left_distance(box1, box2):\n    return (box1[\"bbox\"][2] - box2[\"bbox\"][0]) ** 2 + (\n        box1[\"bbox\"][1] - box2[\"bbox\"][1]\n    ) ** 2\n\nclosest_boxes = []\nfor anchor_det in anchor_dets:\n    # You can use any of the above functions to get the relative position\n    distances = [\n        (relative_det, above_distance(relative_det, anchor_det))\n        for relative_det in relative_dets\n    ]\n    # You must grab the nearest object for each of the anchors. This line will give\n    # you the box directly above the anchor box (or below, left, right depending on\n    # the function used)\n    closest_box = min(distances, key=lambda x: x[1])[0]\n    closest_boxes.append(closest_box)\n\n[end of suggestion 2]\n[end of suggestions]\n\n<count>9<\/count><\/a><\/an><\/an><\/code><\/pre>\n<h3 class=\"wp-block-heading\" id=\"h-step-3-execution\">Step 3: Execution<\/h3>\n<p>The planned steps are executed using the <strong>VisionAgent Library<\/strong> and <strong>Tools Library<\/strong>. This involves loading the image, running models, and generating detection outputs.<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-planner\">Planner<\/h4>\n<p>Based on Claude\u2019s description and the suggestions, I need to detect vegetables both in and around the basket. This is a nested structure problem where I first need to detect the basket and then detect vegetables in relation to it. Let me first get a tool that can detect both the basket and vegetables.<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-example-execution-code\">Example Execution Code<\/h4>\n<pre class=\"wp-block-code\"><code>image = load_image('vegetablebasket.jpg')\nclaude35_vqa('Can you describe this image? What vegetables do you see in and around the basket?', [image])\nsuggestion('How can I detect vegetables in and around the basket in this image?', [image])\nget_tool_for_task('Detect basket and vegetables in the image', [image])<\/code><\/pre>\n<h3 class=\"wp-block-heading\" id=\"h-observation-and-final-output\">Observation and Final Output<\/h3>\n<p>After execution, VisionAgent produces structured results:<\/p>\n<ul class=\"wp-block-list\">\n<li>Detected vegetables categorized inside and outside the basket.<\/li>\n<li>Bounding box coordinates for each detected vegetable.<\/li>\n<li>A deployable AI model that can be used for real-world applications.<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-output\">Output<\/h3>\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1600\" height=\"1007\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958.webp\" alt=\"Output\" class=\"wp-image-220347\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958.webp 1600w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958-300x189.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958-768x483.webp 768w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958-1536x967.webp 1536w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170526.958-150x94.webp 150w\" sizes=\"auto, (max-width: 1600px) 100vw, 1600px\"\/><\/figure>\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1600\" height=\"1007\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275.webp\" alt=\"Output\" class=\"wp-image-220348\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275.webp 1600w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275-300x189.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275-768x483.webp 768w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275-1536x967.webp 1536w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170558.275-150x94.webp 150w\" sizes=\"auto, (max-width: 1600px) 100vw, 1600px\"\/><\/figure>\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1600\" height=\"1007\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946.webp\" alt=\"Output\" class=\"wp-image-220349\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946.webp 1600w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946-300x189.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946-768x483.webp 768w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946-1536x967.webp 1536w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170628.946-150x94.webp 150w\" sizes=\"auto, (max-width: 1600px) 100vw, 1600px\"\/><\/figure>\n<h2 class=\"wp-block-heading\" id=\"h-2-prompt-identify-red-car-in-the-video\">2. Prompt: \u201cIdentify red car in the video\u201d<\/h2>\n<p>Input Video: <a href=\"https:\/\/youtube.com\/shorts\/SyLeJIQkoao?feature=shared\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">Video Link<\/a><\/p>\n<h3 class=\"wp-block-heading\" id=\"h-step-1-conversation-0\">Step 1: Conversation<\/h3>\n<p>I\u2019ll help you write code to identify and track red cars in the video. I\u2019ll use object detection and color-based filtering to accomplish this task.<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-step-2-planner-0\">Step 2: Planner<\/h3>\n<p>I need to first understand what\u2019s in the video and get suggestions on how to identify a red car. I\u2019ll extract some frames and use claude35_vqa to describe the video content, then get suggestions.<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-step-3-execution-0\">Step 3: Execution<\/h3>\n<pre class=\"wp-block-code\"><code>frames = extract_frames_and_timestamps('videoplayback(4).mp4', 1)\nframes = [f[\"frame\"] for f in frames]\nclaude35_vqa('Can you describe this video? Is there a red car in it?', frames[:5])\nsuggestion('How can I identify the red car in the video?', frames[:5])<\/code><\/pre>\n<h3 class=\"wp-block-heading\" id=\"h-observation-0\">Observation<\/h3>\n<pre class=\"wp-block-code\"><code>[claude35_vqa output]\nYes, there is a red car (#21) in this video sequence, which appears to be a NASCAR race clip showing an incident between two cars - a red #21 car and a black\/gold #3 car with Winchester branding. The video shows them making contact around turn 4 of the track, with the cars getting close to each other and potentially trading paint. The sequence is labeled as \"Josh Berry vs Austin Dillon\" and shows what appears to be a tense racing moment between the two drivers. The text overlays indicate \"WOAH TURN 4\", \"CONTACT\", \"BERRY\", \"A LOT\", and \"THERE\" as the sequence progresses.\n[end of claude35_vqa output]\n\n[suggestions]\n\nBased on the user's request to identify the red car in the video and the provided video frames, this is primarily a case of:\n\n1. finding_features_with_video_tracking - The sequence shows a NASCAR race interaction between two cars (Josh Berry and Austin Dillon) where tracking the red #21 car through the turn and contact is key\n2. color - The specific request mentions identifying the \"red car\", making color identification relevant\n3. relative_position - The interaction between the two cars requires understanding their positions relative to each other through the turn and contact\n\nThe other categories don't apply because:\n- This isn't about small text or large images\n- There are no grid patterns or missing elements\n- This isn't about comparing sizes or nested structures\n- While there is some depth perception in racing, the request isn't specifically about determining depth position\n\n\n[suggestion 0]\nYou need to find the color of objects in the image, you can use the following code to help with this task:\n\nimport numpy as np\nimport cv2\n\ncolor_ranges = {\n    \"red_lower\": ((0, 100, 100), (int(179 * 20 \/ 360), 255, 255)),\n    \"orange\": ((int(179 * 21 \/ 360), 100, 100), (int(179 * 50 \/ 360), 255, 255)),\n    \"yellow\": ((int(179 * 51 \/ 360), 100, 100), (int(179 * 70 \/ 360), 255, 255)),\n    \"green\": ((int(179 * 71 \/ 360), 100, 100), (int(179 * 150 \/ 360), 255, 255)),\n    \"cyan\": ((int(179 * 151 \/ 360), 100, 100), (int(179 * 180 \/ 360), 255, 255)),\n    \"blue\": ((int(179 * 181 \/ 360), 100, 100), (int(179 * 265 \/ 360), 255, 255)),\n    \"purple\": ((int(179 * 266 \/ 360), 100, 100), (int(179 * 290 \/ 360), 255, 255)),\n    \"pink\": ((int(179 * 291 \/ 360), 100, 100), (int(179 * 330 \/ 360), 255, 255)),\n    \"red_upper\": ((int(179 * 331 \/ 360), 100, 100), (179, 255, 255)),\n    \"white\": ((0, 0, 200), (179, 25, 255)),\n    \"gray\": ((0, 0, 50), (179, 50, 200)),\n    \"black\": ((0, 0, 0), (179, 255, 30)),\n}\n\ndef get_color(image, color_ranges):\n    hsv_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)\n    detected_colors = {}\n    for color, (lower, upper) in color_ranges.items():\n        upper_range = np.array(upper, dtype=np.uint8)\n        lower_range = np.array(lower, dtype=np.uint8)\n        mask = cv2.inRange(hsv_image, lower_range, upper_range)\n        detected_pixels = cv2.countNonZero(mask)\n        detected_colors[color] = detected_pixels\n\n    if \"red_lower\" in detected_colors and \"red_upper\" in detected_colors:\n        detected_colors[\"red\"] = detected_colors[\"red_lower\"] + detected_colors[\"red_upper\"]\n        del detected_colors[\"red_lower\"]\n        del detected_colors[\"red_upper\"]\n    return sorted(detected_colors, key=detected_colors.get, reverse=True)[0]\n\n[end of suggestion 0]\n[suggestion 1]\nFirst try to solve the problem using a VQA tool before using the tracking approach for a faster and easier solution:\n\nanswer = <vqa tool=\"\" to=\"\" answer=\"\" your=\"\" question=\"\">(\"<your prompt=\"\" here=\"\">\", image)\n\nIf that does not work, you can track the objects in the video and then identify features on those objects. You need to first get a tool that can track objects in a video, and then for each object find another tool to identify the features on the object. You can use the following code to help with this task:\n\ntrack_predictions = <\/your><\/vqa><\/code><\/pre>\n<h4 class=\"wp-block-heading\" id=\"h-planner-1\">Planner<\/h4>\n<p>Based on Claude\u2019s description and the suggestions, I need to track cars in the video and identify the red one. Let me first get a tool for tracking cars in the video.<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-execution-1\">Execution<\/h4>\n<pre class=\"wp-block-code\"><code>get_tool_for_task('Track cars in a NASCAR race video', frames[:5])<\/code><\/pre>\n<h3 class=\"wp-block-heading\" id=\"h-output\">Output<\/h3>\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1207\" height=\"977\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170706.596.webp\" alt=\"Output\" class=\"wp-image-220350\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170706.596.webp 1207w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170706.596-300x243.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170706.596-768x622.webp 768w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/unnamed-2025-02-07T170706.596-150x121.webp 150w\" sizes=\"auto, (max-width: 1207px) 100vw, 1207px\"\/><\/figure>\n<h2 class=\"wp-block-heading\" id=\"h-conclusion\">Conclusion<\/h2>\n<p>VisionAgent is transforming the way developers build, test, and deploy AI-driven vision applications. By automating tedious processes and providing ready-to-use tools, it significantly reduces development time while ensuring high-quality results. Whether you are an AI researcher, a developer, or a business looking to implement computer vision solutions, VisionAgent provides a <strong>fast, flexible, and scalable<\/strong> way to achieve your goals.<\/p>\n<p>With ongoing advancements in AI, VisionAgent is expected to evolve further, incorporating even <strong>more powerful models<\/strong> and expanding its ecosystem to support a <strong>wider range of applications<\/strong>. Now is the perfect time to explore how VisionAgent can enhance your AI-driven vision projects.<\/p>\n<div class=\"border-top py-3 author-info my-4\">\n<div class=\"author-card d-flex align-items-center\">\n<div class=\"flex-shrink-0 overflow-hidden\">\n                                    <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/author\/pankaj9786\/\" class=\"text-decoration-none active-avatar\"><br \/>\n                                                                       <img decoding=\"async\" src=\"https:\/\/av-eks-lekhak.s3.amazonaws.com\/media\/lekhak-profile-images\/converted_image_Lb7Lh0T.webp\" width=\"48\" height=\"48\" alt=\"Pankaj Singh\" loading=\"lazy\" class=\"rounded-circle\"\/><\/p>\n<p>                                <\/a>\n                                <\/div>\n<\/p><\/div>\n<p>                Hi, I am Pankaj Singh Negi &#8211; Senior Content Editor | Passionate about storytelling and crafting compelling narratives that transform ideas into impactful content. I love reading about technology revolutionizing our lifestyle.                 <\/p>\n<\/p><\/div>\n<p>                                       <!-- Free Courses --><\/p>\n<p>                  <!-- Right Side Bar Reading list  --><\/p>\n<div class=\"col-xl-3 col-lg-12 col-md-12 col-sm-12 h-100 side-bar-detail-page z-0 sticky-top\">\n                    <!-- Side bar Reading list --><\/p>\n<div class=\"card border-0\">\n<div class=\"card-body py-0\" id=\"recommeded-list\">\n<div class=\"text-center\"><img decoding=\"async\" src=\"https:\/\/www.analyticsvidhya.com\/wp-content\/themes\/analytics-vidhya\/images\/av_loader.gif\" width=\"100\"\/><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<p>    <!-- Quiz block --><\/p>\n<p>    <!-- FAQ  --><\/p>\n<p>    <!-- Comment Module --><\/p>\n<p>    <!-- Related courses --><\/p>\n<p>    <!-- Write us --><\/p>\n<section class=\"common-style-py\" id=\"writeUs\">\n<div class=\"container-fluid\">\n<div class=\"background-dark-secondary p-5 rounded-3\">\n<div class=\"row aligen-items-center\">\n<div class=\"col-xl-6 col-md-12 col-sm-12\">\n              <a href=\"https:\/\/datahack.analyticsvidhya.com\/blogathon\/\" class=\"text-decoration-none float-end\"><br \/>\n                <img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/www.analyticsvidhya.com\/wp-content\/themes\/analytics-vidhya\/images\/Write-for-us.webp\" alt=\"imag\" width=\"500\" height=\"250\" class=\"img-fluid\"\/><br \/>\n              <\/a>\n            <\/div>\n<\/p><\/div>\n<p>          <!-- Bottom details --><\/p>\n<p>    <!-- \n\n<div class=\"cookies-backdrop\"><\/div>\n\n --><\/p>\n<div class=\"cookies-based-content background-dark-secondary position-fixed bottom-0 w-100 z-3 py-2\" id=\"Cookies\">\n<div class=\"container\">\n<div class=\"row\">\n<div class=\"col-lg-7 col-md-12 col-sm-12\">\n<p class=\"text-white fs-12 fw-light mb-1\">\n                    We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our  <a href=\"https:\/\/www.analyticsvidhya.com\/privacy-policy\" class=\"text-white\">Privacy Policy<\/a> &amp; <a href=\"https:\/\/www.analyticsvidhya.com\/cookies-policy\" class=\"text-white\">Cookies Policy<\/a>.\n                <\/p>\n<p class=\"text-white mt-1 pointer text-decoration-underline fs-16 fw-light mb-0\" data-bs-toggle=\"modal\" data-bs-target=\"#cookiesModal\">Show details<\/p>\n<\/p><\/div>\n<div class=\"col-lg-5 col-md-12 col-sm-12 justify-content-center row align-items-center\">\n<p>\n                    <button type=\"button\" class=\"btn btn-dark btn-dark-primary rounded-3 fs-12 w-100 mb-2 mb-sm-0\" onclick=\"set_cookie()\" aria-label=\"Accept all cookies\">Accept all cookies<\/button>\n                <\/p>\n<p>\n                    <button type=\"button\" class=\"btn btn-dark btn-dark-primary mt-2 rounded-3 fs-12 w-100\" onclick=\"set_cookie()\" aria-label=\"Use Necessary cookies\">Use necessary cookies<\/button>\n                <\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"modal fade\" id=\"cookiesModal\" tabindex=\"-1\" aria-labelledby=\"cookiesModalLabel\" aria-hidden=\"true\">\n<div class=\"modal-dialog modal-lg modal-dialog-centered\">\n<div class=\"modal-content background-dark-secondary p-4\">\n<div class=\"modal-body p-0\">\n<div class=\"tab-content pt-3 px-4\" id=\"myTabContent\">\n<div class=\"tab-pane fade show active py-3\" id=\"consent\" role=\"tabpanel\" aria-labelledby=\"consent-tab\">\n<h6 class=\"text-white\">Cookies<\/h6>\n<p class=\"text-white\">This site uses cookies to ensure that you get the best experience possible. To learn more about how we use cookies, please refer to our <a href=\"https:\/\/www.analyticsvidhya.com\/privacy-policy\" class=\"text-white\" target=\"_blank\">Privacy Policy<\/a> &amp; <a href=\"https:\/\/www.analyticsvidhya.com\/cookies-policy\" class=\"text-white\" target=\"_blank\">Cookies Policy<\/a>.<\/p>\n<\/p><\/div>\n<div class=\"tab-pane fade py-2\" id=\"details\" role=\"tabpanel\" aria-labelledby=\"details-tab\">\n<div class=\"accordion\">\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"headingNecessary\" data-bs-parent=\"#accordionDetails\" id=\"collapseNecessary\">\n<div class=\"accordion\">\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"NecessaryAnalyticsVidhya\" data-bs-parent=\"#collapseNecessary\" id=\"collapseNecessaryAnalyticsVidhya\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>brahmaid<\/h6>\n<p>It is needed for personalizing the website.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>csrftoken<\/h6>\n<p>This cookie is used to prevent Cross-site request forgery (often abbreviated as CSRF) attacks of the website<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>Identityid<\/h6>\n<p>Preserves the login\/logout state of users across the whole site.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>sessionid<\/h6>\n<p>Preserves users&#8217; states across page requests.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"NecessaryGoogle\" data-bs-parent=\"#collapseNecessary\" id=\"collapseNecessaryGoogle\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>g_state<\/h6>\n<p>Google One-Tap login adds this g_state cookie to set the user status on how they interact with the One-Tap modal.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"headingStatistics\" data-bs-parent=\"#accordionDetails\" id=\"collapseStatistics\">\n<div class=\"accordion\">\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"StatisticsMicrosoft\" data-bs-parent=\"#collapseStatistics\" id=\"collapseStatisticsMicrosoft\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>MUID<\/h6>\n<p>Used by Microsoft Clarity, to store and track visits across websites.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_clck<\/h6>\n<p>Used by Microsoft Clarity, Persists the Clarity User ID and preferences, unique to that site, on the browser. This ensures that behavior in subsequent visits to the same site will be attributed to the same user ID.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_clsk<\/h6>\n<p>Used by Microsoft Clarity, Connects multiple page views by a user into a single Clarity session recording.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SRM_I<\/h6>\n<p>Collects user data is specifically adapted to the user or device. The user can also be followed outside of the loaded website, creating a picture of the visitor&#8217;s behavior.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SM<\/h6>\n<p>Use to measure the use of the website for internal analytics<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>CLID<\/h6>\n<p>The cookie is set by embedded Microsoft Clarity scripts. The purpose of this cookie is for heatmap and session recording.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SRM_B<\/h6>\n<p>Collected user data is specifically adapted to the user or device. The user can also be followed outside of the loaded website, creating a picture of the visitor&#8217;s behavior.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"StatisticsGoogle\" data-bs-parent=\"#collapseStatistics\" id=\"collapseStatisticsGoogle\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_gid<\/h6>\n<p>This cookie is installed by Google Analytics. The cookie is used to store information of how visitors use a website and helps in creating an analytics report of how the website is doing. The data collected includes the number of visitors, the source where they have come from, and the pages visited in an anonymous form.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_ga_#<\/h6>\n<p>Used by Google Analytics, to store and count pageviews.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_gat_#<\/h6>\n<p>Used by Google Analytics to collect data on the number of times a user has visited the website as well as dates for the first and most recent visit.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>collect<\/h6>\n<p>Used to send data to Google Analytics about the visitor&#8217;s device and behavior. Tracks the visitor across devices and marketing channels.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>AEC<\/h6>\n<p>cookies ensure that requests within a browsing session are made by the user, and not by other sites.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>G_ENABLED_IDPS<\/h6>\n<p>use the cookie when customers want to make a referral from their gmail contacts; it helps auth the gmail account.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>test_cookie<\/h6>\n<p>This cookie is set by DoubleClick (which is owned by Google) to determine if the website visitor&#8217;s browser supports cookies.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"StatisticsWebengage\" data-bs-parent=\"#collapseStatistics\" id=\"collapseStatisticsWebengage\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_we_us<\/h6>\n<p>this is used to send push notification using webengage.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>WebKlipperAuth<\/h6>\n<p>used by webenage to track auth of webenagage.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"StatisticsLinkedIn\" data-bs-parent=\"#collapseStatistics\" id=\"collapseStatisticsLinkedIn\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>ln_or<\/h6>\n<p>Linkedin sets this cookie to registers statistical data on users&#8217; behavior on the website for internal analytics.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>JSESSIONID<\/h6>\n<p>Use to maintain an anonymous user session by the server.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>li_rm<\/h6>\n<p>Used as part of the LinkedIn Remember Me feature and is set when a user clicks Remember Me on the device to make it easier for him or her to sign in to that device.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>AnalyticsSyncHistory<\/h6>\n<p>Used to store information about the time a sync with the lms_analytics cookie took place for users in the Designated Countries.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>lms_analytics<\/h6>\n<p>Used to store information about the time a sync with the AnalyticsSyncHistory cookie took place for users in the Designated Countries.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>liap<\/h6>\n<p>Cookie used for Sign-in with Linkedin and\/or to allow for the Linkedin follow feature.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>visit<\/h6>\n<p>allow for the Linkedin follow feature.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>li_at<\/h6>\n<p>often used to identify you, including your name, interests, and previous activity.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>s_plt<\/h6>\n<p>Tracks the time that the previous page took to load<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>lang<\/h6>\n<p>Used to remember a user&#8217;s language setting to ensure LinkedIn.com displays in the language selected by the user in their settings<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>s_tp<\/h6>\n<p>Tracks percent of page viewed<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>AMCV_14215E3D5995C57C0A495C55%40AdobeOrg<\/h6>\n<p>Indicates the start of a session for Adobe Experience Cloud<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>s_pltp<\/h6>\n<p>Provides page name value (URL) for use by Adobe Analytics<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>s_tslv<\/h6>\n<p>Used to retain and fetch time since last visit in Adobe Analytics<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>li_theme<\/h6>\n<p>Remembers a user&#8217;s display preference\/theme setting<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>li_theme_set<\/h6>\n<p>Remembers which users have updated their display \/ theme preferences<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"headingPreferences\" data-bs-parent=\"#accordionDetails\" id=\"collapsePreferences\">\n<div class=\"accordion\">\n<p>We do not use cookies of this type.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"headingMarketing\" data-bs-parent=\"#accordionDetails\" id=\"collapseMarketing\">\n<div class=\"accordion\">\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"MarketingGoogle\" data-bs-parent=\"#collapseMarketing\" id=\"collapseMarketingGoogle\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_gcl_au<\/h6>\n<p>Used by Google Adsense, to store and track conversions.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SID<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SAPISID<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>__Secure-#<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>APISID<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>SSID<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>HSID<\/h6>\n<p>Save certain preferences, for example the number of search results per page or activation of the SafeSearch Filter. Adjusts the ads that appear in Google Search.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>DV<\/h6>\n<p>These cookies are used for the purpose of targeted advertising.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>NID<\/h6>\n<p>These cookies are used for the purpose of targeted advertising.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>1P_JAR<\/h6>\n<p>These cookies are used to gather website statistics, and track conversion rates.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>OTZ<\/h6>\n<p>Aggregate analysis of website visitors<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"MarketingFacebook\" data-bs-parent=\"#collapseMarketing\" id=\"collapseMarketingFacebook\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>_fbp<\/h6>\n<p>This cookie is set by Facebook to deliver advertisements when they are on Facebook or a digital platform powered by Facebook advertising after visiting this website.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>fr<\/h6>\n<p>Contains a unique browser and user ID, used for targeted advertising.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"MarketingLinkedIn\" data-bs-parent=\"#collapseMarketing\" id=\"collapseMarketingLinkedIn\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>bscookie<\/h6>\n<p>Used by LinkedIn to track the use of embedded services.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>lidc<\/h6>\n<p>Used by LinkedIn for tracking the use of embedded services.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>bcookie<\/h6>\n<p>Used by LinkedIn to track the use of embedded services.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>aam_uuid<\/h6>\n<p>Use these cookies to assign a unique ID when users visit a website.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>UserMatchHistory<\/h6>\n<p>These cookies are set by LinkedIn for advertising purposes, including: tracking visitors so that more relevant ads can be presented, allowing users to use the &#8216;Apply with LinkedIn&#8217; or the &#8216;Sign-in with LinkedIn&#8217; functions, collecting information about how visitors use the site, etc.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>li_sugr<\/h6>\n<p>Used to make a probabilistic match of a user&#8217;s identity outside the Designated Countries<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"MarketingMicrosoft\" data-bs-parent=\"#collapseMarketing\" id=\"collapseMarketingMicrosoft\">\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>MR<\/h6>\n<p>Used to collect information for analytics purposes.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"card mb-3\">\n<div class=\"card-body pb-1\">\n<h6>ANONCHK<\/h6>\n<p>Used to store session ID for a users session to ensure that clicks from adverts on the Bing search engine are verified for reporting purposes and for personalisation<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"accordion-item\">\n<div class=\"accordion-body accordion-collapse collapse\" aria-labelledby=\"headingUnclassNameified\" data-bs-parent=\"#accordionDetails\" id=\"collapseUnclassNameified\">\n<div class=\"accordion\">\n<p>We do not use cookies of this type.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<p class=\"mt-3 ms-2 text-white\">Cookie declaration last updated on 24\/03\/2023 by Analytics Vidhya.<\/p>\n<\/p><\/div>\n<div class=\"tab-pane fade py-2\" id=\"about\" role=\"tabpanel\" aria-labelledby=\"about-tab\">\n<p class=\"text-white fs-18 fs-light\">Cookies are small text files that can be used by websites to make a user&#8217;s experience more efficient. The law states that we can store cookies on your device if they are strictly necessary for the operation of this site. For all other types of cookies, we need your permission. This site uses different types of cookies. Some cookies are placed by third-party services that appear on our pages. Learn more about who we are, how you can contact us, and how we process personal data in our <a href=\"https:\/\/www.analyticsvidhya.com\/privacy-policy\" class=\"text-white\" target=\"_blank\">Privacy Policy<\/a>.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<div class=\"modal login-modal shadow\" aria-hidden=\"true\" aria-labelledby=\"emailModalLabel\" id=\"emailModal\" data-bs-keyboard=\"false\" data-bs-backdrop=\"static\" tabindex=\"-1\">\n<div class=\"modal-dialog modal-dialog-centered\">\n<div class=\"modal-content background-dark-primary shadow-sm rounded-4 p-4\">\n<div class=\"modal-body p-0 pt-5\">\n<div class=\"d-flex\">\n                <svg data-bs-toggle=\"modal\" data-bs-target=\"#loginModal\" class=\"me-2 backBtn\" width=\"24\" height=\"24\" viewbox=\"0 0 24 24\" fill=\"none\">\n                    <path d=\"M19 12H5M5 12L12 19M5 12L12 5\" stroke=\"white\" strokewidth=\"2\" strokelinecap=\"round\" strokelinejoin=\"round\"\/>\n                <\/svg><\/p>\n<h2 class=\"fs-20 text-white mb-4\">Enter email address to continue<\/h2>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n<div class=\"modal login-modal shadow\" id=\"otpModal\" aria-labelledby=\"loginOtpModalLabel\" tabindex=\"-1\" data-bs-keyboard=\"false\" data-bs-backdrop=\"static\" aria-hidden=\"true\">\n<div class=\"modal-dialog modal-dialog-centered\">\n<div class=\"modal-content background-dark-primary shadow-sm rounded-4 p-4\">\n<div class=\"modal-body p-0 pt-5\">\n<p class=\"blue pointer \" id=\"resendOtpBtn\">Resend OTP<\/p>\n<p class=\"text-dark-tertiary d-none\">Resend OTP in <span class=\"blue\" id=\"resentOtpSecond\">45s<\/span><\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div><\/div>\n<\/div>\n<\/section>\n<\/div>\n\n","protected":false},"excerpt":{"rendered":"<p>Model Recall Precision F1 Score Landing AI 77.0% 82.6% 79.7% (highest) Microsoft Florence-2 43.4% 36.6% 39.7% Google OWLv2 81.0% 29.5% 43.2% Alibaba Qwen2.5-VL-7B-Instruct 26.0% 54.0% 35.1% 4. Key Takeaways Landing AI\u2019s Agentic Object Detection achieved the highest F1 Score (79.7%), meaning it balances precision and recall better than the others. Google OWLv2 had the highest [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":74019,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[12033],"tags":[3481,38253,11205,35522,8767,38254],"dealstore":[],"offerexpiration":[],"class_list":["post-74018","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-analytics","tag-andrew","tag-ngs","tag-solutions","tag-streamlining","tag-vision","tag-visionagent"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fivemor.com\/?p=74018\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network\" \/>\n<meta property=\"og:description\" content=\"Model Recall Precision F1 Score Landing AI 77.0% 82.6% 79.7% (highest) Microsoft Florence-2 43.4% 36.6% 39.7% Google OWLv2 81.0% 29.5% 43.2% Alibaba Qwen2.5-VL-7B-Instruct 26.0% 54.0% 35.1% 4. Key Takeaways Landing AI\u2019s Agentic Object Detection achieved the highest F1 Score (79.7%), meaning it balances precision and recall better than the others. Google OWLv2 had the highest [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fivemor.com\/?p=74018\" \/>\n<meta property=\"og:site_name\" content=\"Som2ny Network\" \/>\n<meta property=\"article:published_time\" content=\"2025-02-07T16:47:50+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"872\" \/>\n\t<meta property=\"og:image:height\" content=\"473\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"18 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fivemor.com\/?p=74018#article\",\"isPartOf\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018\"},\"author\":{\"name\":\"admin\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371\"},\"headline\":\"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions\",\"datePublished\":\"2025-02-07T16:47:50+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018\"},\"wordCount\":2006,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fivemor.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018#primaryimage\"},\"thumbnailUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp\",\"keywords\":[\"Andrew\",\"Ngs\",\"Solutions\",\"Streamlining\",\"vision\",\"VisionAgent\"],\"articleSection\":[\"Analytics\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fivemor.com\/?p=74018#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fivemor.com\/?p=74018\",\"url\":\"https:\/\/fivemor.com\/?p=74018\",\"name\":\"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network\",\"isPartOf\":{\"@id\":\"https:\/\/fivemor.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018#primaryimage\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018#primaryimage\"},\"thumbnailUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp\",\"datePublished\":\"2025-02-07T16:47:50+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/fivemor.com\/?p=74018#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fivemor.com\/?p=74018\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/?p=74018#primaryimage\",\"url\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp\",\"contentUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp\",\"width\":872,\"height\":473},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fivemor.com\/?p=74018#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/fivemor.com\/?bp_activities=1\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fivemor.com\/#website\",\"url\":\"https:\/\/fivemor.com\/\",\"name\":\"Som2ny Network\",\"description\":\"Daily Deals\",\"publisher\":{\"@id\":\"https:\/\/fivemor.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fivemor.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/fivemor.com\/#organization\",\"name\":\"Som2ny Network\",\"url\":\"https:\/\/fivemor.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png\",\"contentUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png\",\"width\":300,\"height\":86,\"caption\":\"Som2ny Network\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png\",\"caption\":\"admin\"},\"sameAs\":[\"https:\/\/fivemor.com\"],\"url\":\"https:\/\/fivemor.com\/?author=1\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fivemor.com\/?p=74018","og_locale":"en_US","og_type":"article","og_title":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network","og_description":"Model Recall Precision F1 Score Landing AI 77.0% 82.6% 79.7% (highest) Microsoft Florence-2 43.4% 36.6% 39.7% Google OWLv2 81.0% 29.5% 43.2% Alibaba Qwen2.5-VL-7B-Instruct 26.0% 54.0% 35.1% 4. Key Takeaways Landing AI\u2019s Agentic Object Detection achieved the highest F1 Score (79.7%), meaning it balances precision and recall better than the others. Google OWLv2 had the highest [&hellip;]","og_url":"https:\/\/fivemor.com\/?p=74018","og_site_name":"Som2ny Network","article_published_time":"2025-02-07T16:47:50+00:00","og_image":[{"width":872,"height":473,"url":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp","type":"image\/webp"}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"Written by":"admin","Est. reading time":"18 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fivemor.com\/?p=74018#article","isPartOf":{"@id":"https:\/\/fivemor.com\/?p=74018"},"author":{"name":"admin","@id":"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371"},"headline":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions","datePublished":"2025-02-07T16:47:50+00:00","mainEntityOfPage":{"@id":"https:\/\/fivemor.com\/?p=74018"},"wordCount":2006,"commentCount":0,"publisher":{"@id":"https:\/\/fivemor.com\/#organization"},"image":{"@id":"https:\/\/fivemor.com\/?p=74018#primaryimage"},"thumbnailUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp","keywords":["Andrew","Ngs","Solutions","Streamlining","vision","VisionAgent"],"articleSection":["Analytics"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fivemor.com\/?p=74018#respond"]}]},{"@type":"WebPage","@id":"https:\/\/fivemor.com\/?p=74018","url":"https:\/\/fivemor.com\/?p=74018","name":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions - Som2ny Network","isPartOf":{"@id":"https:\/\/fivemor.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/fivemor.com\/?p=74018#primaryimage"},"image":{"@id":"https:\/\/fivemor.com\/?p=74018#primaryimage"},"thumbnailUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp","datePublished":"2025-02-07T16:47:50+00:00","breadcrumb":{"@id":"https:\/\/fivemor.com\/?p=74018#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fivemor.com\/?p=74018"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/?p=74018#primaryimage","url":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp","contentUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/VisionAgent-.webp.webp","width":872,"height":473},{"@type":"BreadcrumbList","@id":"https:\/\/fivemor.com\/?p=74018#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/fivemor.com\/?bp_activities=1"},{"@type":"ListItem","position":2,"name":"Andrew Ng\u2019s VisionAgent: Streamlining Vision AI Solutions"}]},{"@type":"WebSite","@id":"https:\/\/fivemor.com\/#website","url":"https:\/\/fivemor.com\/","name":"Som2ny Network","description":"Daily Deals","publisher":{"@id":"https:\/\/fivemor.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fivemor.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/fivemor.com\/#organization","name":"Som2ny Network","url":"https:\/\/fivemor.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/#\/schema\/logo\/image\/","url":"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png","contentUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png","width":300,"height":86,"caption":"Som2ny Network"},"image":{"@id":"https:\/\/fivemor.com\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371","name":"admin","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png","caption":"admin"},"sameAs":["https:\/\/fivemor.com"],"url":"https:\/\/fivemor.com\/?author=1"}]}},"_links":{"self":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts\/74018","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=74018"}],"version-history":[{"count":0,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts\/74018\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/media\/74019"}],"wp:attachment":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=74018"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=74018"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=74018"},{"taxonomy":"dealstore","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fdealstore&post=74018"},{"taxonomy":"offerexpiration","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fofferexpiration&post=74018"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}