# 1st Place Solution to ECCV-TAO-2020: Detect and Represent Any Object for Tracking

arXiv: https://arxiv.org/abs/2101.08040 · HTML: https://mlsanity.com/paper/2101.08040/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 15 references that are in the map.
- First posted on arXiv 20 Jan 2021.

## Builds on (most cited first, 13 of 13)

- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [MOT16: A Benchmark for Multi-Object Tracking](https://mlsanity.com/paper/1603.00831/) (2016)
- [GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild](https://mlsanity.com/paper/1810.11981/) (2018)
- [SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks](https://mlsanity.com/paper/1812.11703/) (2018)
- [Bag of Tricks and A Strong Baseline for Deep Person Re-identification](https://mlsanity.com/paper/1903.07071/) (2019)
- [Tracking without bells and whistles](https://mlsanity.com/paper/1903.05625/) (2019)
- [DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution](https://mlsanity.com/paper/2006.02334/) (2020)
- [A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification](https://mlsanity.com/paper/1906.08332/) (2019)
- [Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax](https://mlsanity.com/paper/2006.10408/) (2020)
- [TAO: A Large-Scale Benchmark for Tracking Any Object](https://mlsanity.com/paper/2005.10356/) (2020)
- [YouTube-BoundingBoxes: A Large High-Precision Human-Annotated Data Set for Object Detection in Video](https://mlsanity.com/paper/1702.00824/) (2017)
- [Deep Cosine Metric Learning for Person Re-Identification](https://mlsanity.com/paper/1812.00442/) (2018)

## Built on by (most cited first, 17 of 17)

- [Global Tracking Transformers](https://mlsanity.com/paper/2203.13250/) (2022)
- [QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking](https://mlsanity.com/paper/2210.06984/) (2022)
- [OVTrack: Open-Vocabulary Multiple Object Tracking](https://mlsanity.com/paper/2304.08408/) (2023)
- [Opening up Open-World Tracking](https://mlsanity.com/paper/2104.11221/) (2021)
- [Matching Anything by Segmenting Anything](https://mlsanity.com/paper/2406.04221/) (2024)
- [Tracking Every Thing in the Wild](https://mlsanity.com/paper/2207.12978/) (2022)
- [NetTrack: Tracking Highly Dynamic Objects with a Net](https://mlsanity.com/paper/2403.11186/) (2024)
- [City-Scale Multi-Camera Vehicle Tracking Guided by Crossroad Zones](https://mlsanity.com/paper/2105.06623/) (2021)
- [Video OWL-ViT: Temporally-consistent open-world localization in video](https://mlsanity.com/paper/2308.11093/) (2023)
- [TAO-Amodal: A Benchmark for Tracking Any Object Amodally](https://mlsanity.com/paper/2312.12433/) (2023)
- [SLAck: Semantic, Location, and Appearance Aware Open-Vocabulary Tracking](https://mlsanity.com/paper/2409.11235/) (2024)
- [VOVTrack: Exploring the Potentiality in Videos for Open-Vocabulary Object Tracking](https://mlsanity.com/paper/2410.08529/) (2024)
- [OCTrack: Benchmarking the Open-Corpus Multi-Object Tracking](https://mlsanity.com/paper/2407.14047/) (2024)
- [Reliable Multi-Object Tracking in the Presence of Unreliable Detections](https://mlsanity.com/paper/2112.08345/) (2021)
- [The Progression of Transformers from Language to Vision to MOT: A Literature Review on Multi-Object Tracking with Transformers](https://mlsanity.com/paper/2406.16784/) (2024)
- [Lattice Boltzmann Model for Learning Real-World Pixel Dynamicity](https://mlsanity.com/paper/2509.16527/) (2025)
- [COVTrack++: Learning Open-Vocabulary Multi-Object Tracking from Continuous Videos via a Synergistic Paradigm](https://mlsanity.com/paper/2603.24016/) (2026)

## Abstract (arXiv)

We extend the classical tracking-by-detection paradigm to this tracking-any-object task. Solid detection results are first extracted from TAO dataset. Some state-of-the-art techniques like BAlanced-Group Softmax (BAGSciteli2020overcoming) and DetectoRSciteqiao2020detectors are integrated during detection. Then we learned appearance features to represent any object by training feature learning networks. We ensemble several models for improving detection and feature representation. Simple linking strategies with most similar appearance features and tracklet-level post association module are finally applied to generate final tracking results. Our method is submitted as AOA on the challenge website. Code is available at https://github.com/feiaxyt/WinnerECCV20TAO.