一、Perception

1. D³Fields: Dynamic 3D Descriptor Fields for Zero-Shot Generalizable Rearrangement (CoRL-2024)
   [paper] [code]

2. UAD: Unsupervised Affordance Distillation for Generalization in Robotic Manipulation (ICRA-2025)
   [paper] [code]

3. Learning Affordance Grounding from Exocentric Images (CVPR-2022)
   [paper] [code]

4. Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation (CVPR-2025)
   [paper] [code]

5. AffordanceLLM: Grounding Affordance from Vision Language Models (CVPR-2024)
   [paper] [code]

6. DINOv3
   [paper] [code]

7. SAM 3: Segment Anything with Concepts
   [paper] [code]

[Nan Xue]
[Yuxi Xiao]

Depth Completion

1. D³RoMa: Disparity Diffusion-based Depth Sensing for Material-Agnostic Robotic Manipulation (CoRL-2024)
   [paper] [code]

2. Prompting Depth Anything for 4K Resolution Accurate Metric Depth Estimation (CVPR-2025)
   [paper] [code]

3. Masked Depth Modeling for Spatial Perception
   [paper] [code]

4. Manipulation as in Simulation: Enabling Accurate Geometry Perception in Robots
   [paper] [code]

Object Tracking

1. SpatialTracker: Tracking Any 2D Pixels in 3D Space (CVPR-2024)
   [paper] [code]

2. SpatialTrackerV2: 3D Point Tracking Made Easy (ICCV-2025)
   [paper] [code]

3. Tracking Any Point
   [Link]

4. CoTracker: It is Better to Track Together (ECCV-2024)
   [paper] [code]

5. PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking (ICCV-2023)
   [paper] [code]

二、Robotic Manipulation

1. Diffusion Policy: Visuomotor Policy Learning via Action Diffusion (RSS-2023/IJRR-2024)
   [paper] [code]

2. 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations (RSS-2024)
   [paper] [code]

3. Generalizable Humanoid Manipulation with 3D Diffusion Policies (IROS-2025)
   [paper] [code]

4. Motion Before Action: Diffusing Object Motion as Manipulation Condition (RA-L-2025/ICRA-2026)
   [paper] [code]

5. GenDP: 3D Semantic Fields for Category-Level Generalizable Diffusion Policy (CoRL-2024)
   [paper] [code]

6. ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation (CoRL-2024)
   [paper] [code]

7. VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models (CoRL-2023)
   [paper] [code]

8. Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes (RA-L-2023)
   [paper] [code]

9. Diff-DAgger: Uncertainty Estimation with Diffusion Policy for Robotic Manipulation(ICRA-2025)
   [paper] [code]

[Wenlong Huang]
[Yanjie Ze]
[Yixuan Wang]

Vision Language Action

[Link]

三、Framework

1. LeRobot
2. Isaac Sim
3. Isaac Lab
4. ROS2

四、Technical Roadmap

[Lumina-Embodied-AI-Guide]