Computer-Vision

1. COCO (Common Objects in Context) Description: COCO is a large-scale dataset containing images of objects in complex scenes with annotations for object detection, segmentation, and keypoints. Download: COCO Dataset License: Creative Commons Attribution 4.0 License
2. Pascal VOC (Visual Object Classes) Description: The Pascal VOC dataset provides images with bounding box annotations for various object classes, including people. Download: Pascal VOC Dataset License: VOC license
3. INRIA Person Dataset Description: The INRIA Person Dataset is specifically focused on pedestrian detection with annotated images of people. Download: INRIA Person Dataset License: Research-only license
4. Open Images Dataset Description: Open Images is a dataset with annotations for object detection, segmentation, and visual relationship detection, including a significant number of people annotations. Download: Open Images Dataset License: Creative Commons Attribution 4.0 International License
5. CrowdHuman Dataset Description: CrowdHuman is a dataset for detecting human instances in crowded scenes with comprehensive annotations. Download: CrowdHuman Dataset License: Research use only
6. WIDER Pedestrian Description: WIDER Pedestrian is a dataset for pedestrian detection in various scenes with bounding box annotations. Download: WIDER Pedestrian Dataset License: Academic use only