Resources

Datasets

  • The NeuralGrasps dataset is available here.
  • The FewSOL dataset is available here.
  • The DexYCB dataset is available here.
  • The Tabletop Object Dataset (TOD) for unseen object instance segmentation is available here (34G).
  • The YCB-Video dataset is available here.
  • The Stanford Online Products dataset is available here.
  • The ObjectNet3D dataset is available here.
  • The Multiview Tracking dataset is available here.
  • The PASCAL3D+ dataset is available here.

Software and Source Code

Robot Manipulation
  • The source code of NeuralGrasps: Learning Implicit Representations for Grasps of Multiple Robotic Hands is available here.
  • The source code of Hierarchical Policies for Cluttered-Scene Grasping with Latent Plans is available here.
  • The source code of GA-DDPG: Goal-Auxiliary Actor-Critic for 6D Robotic Grasping with Point Clouds is available here.
  • The source code of OMG-Planner: Manipulation Trajectory Optimization with Online Grasp Synthesis and Selection is available here.
Robot Navigation
  • The source code of Learning Composable Behavior Embeddings for Long-horizon Visual Navigation is available here.
  • The source code of Neural Autonomous Navigation with Riemannian Motion Policy is available here.
VR/AR
  • A data capture tool for Microsoft HoloLens 2 is available here.
Object Segmentation
  • The source code of RICE: Refining Instance Masks in Cluttered Environments with Graph Neural Networks is available here.
  • The source code of Unseen Object Instance Segmentation for Robotic Environments is available here.
  • The source code of Learning RGB-D Feature Embeddings for Unseen Object Instance Segmentation is available here.
Object Pose Estimation
  • The source code of LatentFusion: End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation is available here.
  • The source code of PoseRBPF: A Rao-Blackwellized Particle Filter for 6D Object Pose Tracking is available here.
  • The source code DOPE: Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objects is available here.
  • The source code of DeepIM: Deep Iterative Matching for 6D Pose Estimation is available here.
  • The source code of PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes is available here.
  • The source code of Data-Driven 3D Voxel Patterns for Object Category Recognition is available here.
  • The source code of Object Detection by 3D Aspectlets and Occlusion Reasoning is available here.
Object Tracking
  • The source code of Learning to Track: Online Multi-Object Tracking by Decision Making is available here.
  • The source code of Monocular Multiview Object Tracking with 3D Aspect Parts is available here.
Semantic Mapping
  • The source code of DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks is available here.
Deep Metric Learning
  • The source code of Deep Metric Learning via Lifted Structured Feature Embedding is available here.