🚗 Predict multi-agent trajectories for autonomous vehicles using advanced models on the Argoverse 2 dataset, enhancing motion forecasting capabilities.

Numpy-only dynamic object removal for LiDAR point clouds. 3D box crop + temporal filtering, ROS2 realtime node, public AV2 demos.

[ICCV 2025] DONUT: A Decoder-Only Model for Trajectory Prediction

Argoverse 2: Next generation datasets for self-driving perception and forecasting.

"Official code for TCRAT-Pred: Multi-Agents Trajectory Prediction for Autonomous Vehicles with Multi-Modal Predictions (IEEE CogInfoCom 2024)"

[RAL 2025]Multi-Agent Trajectory Prediction with Difficulty-Guided Feature Enhancement Network

This is useful to debug the Argoverse V1.1 dataset.

[RAL 2025]Multi-Agent Trajectory Prediction with Difficulty-Guided Feature Enhancement Network-xinguipeng

Using PyTorch to predict where cars, cyclists, and pedestrians will go.

The official implementation of "Hierarchical Latent Structure for Multi-Modal Vehicle Trajectory Forecasting" presented in ECCV2022.

[CVPR 2024 MULA]Official PyTorch Implementation of "LAformer: Trajectory Prediction for Autonomous Driving with Lane-Aware Scene Constraints"

This repository contains our work on a comprehensive investigation on motion prediction for Autonomous Vehicles using the PowerBEV framework and a Multi-Camera setup. Validated trajectory forecasting capabilities on the NuScenes, Woven and Argoverse datasets and identified challenges in model generalization across these datasets.

Official github for Delay-adaptive Detector

[CVPR 2023] Query-Centric Trajectory Prediction

TrafficBots: Towards World Models for Autonomous Driving Simulation and Motion Prediction. ICRA 2023. You may also want to check out the updated version: https://github.com/zhejz/TrafficBotsV1.5

Long Range 3D Perception - VoxelNeXt (CVPR 2023)

Trajectory Prediction with Local Self-Attentive Contexts for Autonomous Driving (NeurIPS 2020)

Monocular depth estimation from ArgoAI's Lidar based Depth dataset - Depth predictions up-to 200m

Seperate dual lidar lasers and load the intensity and ring-numbers for better control over lidar data. Using Argoverse dataset.

ROS API and utils for Argoverse dataset