MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors

This repository is an official implementation of MOTRv2.

Introduction

TL; DR. MOTRv2 improve MOTR by utilizing YOLOX to provide detection prior.

Abstract. In this paper, we propose MOTRv2, a simple yet effective pipeline to bootstrap end-to-end multi-object tracking with a pretrained object detector. Existing end-to-end methods, e.g. MOTR and TrackFormer are inferior to their tracking-by-detection counterparts mainly due to their poor detection performance. We aim to improve MOTR by elegantly incorporating an extra object detector. We first adopt the anchor formulation of queries and then use an extra object detector to generate proposals as anchors, providing detection prior to MOTR. The simple modification greatly eases the conflict between joint learning detection and association tasks in MOTR. MOTRv2 keeps the end-to-end feature and scales well on large-scale benchmarks. MOTRv2 achieves the top performance (73.4% HOTA) among all existing methods on the DanceTrack dataset. Moreover, MOTRv2 reaches state-of-the-art performance on the BDD100K dataset. We hope this simple and effective pipeline can provide some new insights to the end-to-end MOT community.

News

• 2023.02.28 MOTRv2 is accepted to CVPR 2023.
• 2022.11.18 MOTRv2 paper is available on arxiv.
• 2022.10.27 Our DanceTrack challenge tech report is released [arxiv] [ECCVW Challenge].
• 2022.10.05 MOTRv2 achieved the 1st place in the 1st Multiple People Tracking in Group Dance Challenge.

Main Results

DanceTrack

Visualization

SORT-like SoTA	MOTRv2

Installation

The codebase is built on top of Deformable DETR and MOTR.

Requirements

• Install pytorch using conda (optional)

  conda create -n motrv2 python=3.7
  conda activate motrv2
  conda install pytorch=1.8.1 torchvision=0.9.1 cudatoolkit=10.2 -c pytorch

• Other requirements

  pip install -r requirements.txt

• Build MultiScaleDeformableAttention

  cd ./models/ops
  sh ./make.sh

Usage

Dataset preparation

1. Download YOLOX detection from here.
2. Please download DanceTrack and CrowdHuman and unzip them as follows:

/data/Dataset/mot
├── crowdhuman
│   ├── annotation_train.odgt
│   ├── annotation_trainval.odgt
│   ├── annotation_val.odgt
│   └── Images
├── DanceTrack
│   ├── test
│   ├── train
│   └── val
├── det_db_motrv2.json

You may use the following command for generating crowdhuman trainval annotation:

cat annotation_train.odgt annotation_val.odgt > annotation_trainval.odgt

Training

You may download the coco pretrained weight from Deformable DETR (+ iterative bounding box refinement), and modify the --pretrained argument to the path of the weight. Then training MOTR on 8 GPUs as following:

./tools/train.sh configs/motrv2.args

Inference on DanceTrack Test Set

# run a simple inference on our pretrained weights
./tools/simple_inference.sh ./motrv2_dancetrack.pth

# Or evaluate an experiment run
# ./tools/eval.sh exps/motrv2/run1

# then zip the results
zip motrv2.zip tracker/ -r

Acknowledgements

• MOTR
• ByteTrack
• YOLOX
• OC-SORT
• DanceTrack
• BDD100K