Current task types
May 20, 2025 ยท View on GitHub
This list shows how we reformulating existing visual perception tasks. We are working on it to support more task types.
Detection
- Visual Grounding
- Object Detection
- 2D Object Detection
- Small Object Detection
- Defect Detection
- Face Detection
- License Plate Detection
- Anomaly Detection
- Human Detection
- Surgical Tool Detection
- Dense Object Detection
- Open World Object Detection
- Zero-Shot Object Detection
- Animal Action Recognition
- Robotic Grasping
- Object Localization
- Hand Detection
- Visual Relationship Detection
- Open Vocabulary Object Detection
- Oriented Object Detection
- Object Detection in Indoor Scenes
- Object Detection in Aerial Images
- Person Search
- Object Recognition
Segmentation
- Semantic Segmentation
- Instance Segmentation
- Lane Detection
- 2D Semantic Segmentation
- Medical Image Segmentation
- Human Part Segmentation
- Action Segmentation
- Video Object Segmentation
- Referring Expression Segmentation
- Saliency Detection
- Salient Object Detection
- The Semantic Segmentation of Remote Sensing Imagery
- Crack Segmentation
- Action Unit Detection
- RGB Salient Object Detection
- Boundary Detection
- Crack Segmentation for Infrastructure
- Surgical Tool Segmentation
Counting
- Object Counting
- Crowd Counting
- Density Estimation
- Pedestrian Detection
- Crowd Estimation in Dense Scenes
- Traffic Counting in Surveillance