ViT-Up: Faithful Feature Upsampling for Vision Transformers

STaR: Scalable Task-Conditioned Retrieval for Long-Horizon Multimodal Robot Memory

A structured reading list on Vision-Language-Action (VLA) models — from diffusion/flow matching foundations through state-of-the-art robot foundation model architectures to data scaling, RL fine-tu…

Open Source Robotic Arm for All Developers

Official implementation of paper "VLM³: Vision Language Models Are Native 3D Learners".

[ICCV 2025] AGO: Adaptive Grounding for Open World 3D Occupancy Prediction

[AAAI 2025] Official implementation of the paper "EOV-Seg: Efficient Open-Vocabulary Panoptic Segmentation"

[CVPR 2026] An Instance-Centric Panoptic Occupancy Prediction Benchmark for Autonomous Driving

Official Implementation of "CAT-Seg🐱: Cost Aggregation for Open-Vocabulary Semantic Segmentation"

WildOS: Open-Vocabulary Object Search in the Wild

IROS/ISVC 2023: Multimodal Dataset for Localization, Mapping and Crop Monitoring in Citrus Tree Farms

SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning

Geometric Traversability using Self-Supervised with Anomaly Classification @ ICRA'26

SocraticKG: Knowledge Graph Construction via QA-Driven Fact Extraction

Awesome papers and code about Multi-Camera 3D Occupancy Prediction, such as TPVFormer, SurroundOcc, PanoOcc, OccFormer, FB-OCC, SelfOcc, COTR, SparseOcc, GaussianFormer, GaussianOcc, STCOcc, OccMam…

Official Implementation of Upsample Anything: A Simple and Hard to Beat Baseline for Feature Upsampling

[TCSVT] DAOcc: 3D Object Detection Assisted Multi-Sensor Fusion for 3D Occupancy Prediction

[RA-L '24] Learning self-supervised traversability with navigation experiences of mobile robots: A risk-aware self-training approach

Self-Supervised LiDAR Ground Segmentation [CVPR 2026]

EllipseLIO is an Adaptive LiDAR Inertial Odometry Approach with an Ellipsoid Representation

[RSS 2026] Learning Point Cloud Geometry as a Statistical Manifold: Theory and Practice

Convert nuscenes data to ros2bag file format

[ECCV 2026] Any2Full: Prompting Depth Anything for Depth Completion in One Stage

Patchwork++: Fast and robust ground segmentation method for 3D LiDAR scans. @ IROS'22

Scene as Occupancy and Reconstruction: A Comprehensive Dataset for Unstructured Scene Understanding

Forest3D is a comprehensive toolkit that generates realistic forest environments for Gazebo robotics simulation by processing Blender assets and DEM terrain data through automated procedural placem…

[ICCV2025] Accelerate 3D Object Detection Models via Zero-Shot Attention Key Pruning

[ECCV 2024] Occupancy as Set of Points

[CVPR 2025] GaussTR: Foundation Model-Aligned Gaussian Transformer for Self-Supervised 3D Spatial Understanding