The open source implementation of 'Offline Tracking with Object Permanence', which aims to recover the occluded vehicle trajectories and reduce the identity switches caused by occlusions.
-
Updated
Jan 3, 2024 - Python
The open source implementation of 'Offline Tracking with Object Permanence', which aims to recover the occluded vehicle trajectories and reduce the identity switches caused by occlusions.
Music conditioned dance prediction
We propose to utilize Transformers as the architectural choice for predicting human motion.
Motion Prediction for Self-driving cars using Polyline-LSTM-Transformer based architecture (based on Multipath++).
Vehicle-Pedestrian Motion Prediction using Classification
Code for "Towards Explainable Multi-modal Motion Prediction using Graph Representations"
Code for "FloMo: Tractable Motion Prediction with Normalizing Flows", Schöller et al., International Conference on Intelligent Robots and Systems (IROS), 2021
Reinforcement learning agent that actively learns physical properties (motion mechanisms) in a 2D simulated domain.
[CVPR 2023] PF-Track (3D MOT for Autonomous Driving)
Deep learning algorithm to perform motion prediction on vehicles as part of an autonomous car software stack.
TrafficBots V1.5: TrafficBots + HPTR. 3rd place solution for Waymo Open Sim Agent Challenge 2024.
[ECCV 2024] Accelerating Online Mapping and Behavior Prediction via Direct BEV Feature Attention
Code and results/visualizations for the paper "A Neural Temporal Model for Human Motion Prediction", CVPR 2019
Pytorch implementation for the paper: "OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment" [Neurocomputing 2024]
Implementation of "MotionCNN: A Strong Baseline for Motion Prediction in Autonomous Driving" for Waymo Open Motion Dataset
Official implementation of dual quaternion transformations as described in the paper "Pose Representations for Deep Skeletal Animation".
Code for: "Skeleton-Graph: Long-Term 3D Motion Prediction From 2D Observations Using Deep Spatio-Temporal Graph CNNs", ICCV2021 Workshops
Deep learning models for self-driving vehicles to predict other car/cyclist/pedestrian (called "agent")'s motion.
Full conference version of PVT++: A Simple End-to-End Latency-Aware Visual Tracking Framework
Add a description, image, and links to the motion-prediction topic page so that developers can more easily learn about it.
To associate your repository with the motion-prediction topic, visit your repo's landing page and select "manage topics."