Name: Ruochen Li
Type: User
Company: Durham University
Bio: My name Ruochen Li, currently a PhD student in Durham University
Location: Durham, England
Ruochen Li's Projects
Efficient AI Backbones including GhostNet, TinyNet and TNT, developed by Huawei Noah's Ark Lab.
Official Code for "EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecasting (ICCV 2023)"
Introductory tutorials for trajectory generation.
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
[ICCV'2023] Forecast-MAE: Self-supervised Pre-training for Motion Forecasting with Masked Autoencoders
Implementation of Gaussian Mixture Variational Autoencoder (GMVAE) for Unsupervised Clustering
Official Code for "Learning Pedestrian Group Representations for Multi-modal Trajectory Prediction (ECCV 2022)"
Official Code for "A Set of Control Points Conditioned Pedestrian Trajectory Prediction (AAAI 2023)"
[CVPR 2022] HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction
程序员在家做饭方法指南。Programmer's guide about how to cook at home (Chinese only).
Our ECCV 2022 paper Human Trajectory Prediction via Neural Social Physics
Crawl & visualize ICLR papers and reviews.
This repository contains all the codes share with a nice person~
An implementation of LaneGCN (Learning Lane Graph Representations for Motion Forecasting)
Tips for Writing a Research Paper using LaTeX
Implement layer normalization GRU in pytorch
PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡
This is the official implementation for AAAI-23 Oral paper "Are Transformers Effective for Time Series Forecasting?"
Official PyTorch code for "Remember Intentions: Retrospective-Memory-based Trajectory Prediction".
🏀 Visualization of NBA games from raw SportVU data logs
Scalable and user friendly neural :brain: forecasting algorithms.
Human Trajectory Prediction Dataset Benchmark (ACCV 2020)
Code for "Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Plans" https://arxiv.org/abs/2001.00735
The visualization of annotation files for different pedestrian datasets.
POWERBEV, a novel and elegant vision-based end-to-end framework that only consists of 2D convolutional layers to perform perception and forecasting of multiple objects in BEVs.