# TAPTRv3: Spatial and Temporal Context Foster Robust Tracking of Any Point in Long Video

arXiv: https://arxiv.org/abs/2411.18671 · HTML: https://mlsanity.com/paper/2411.18671/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 45 of its 60 references that are in the map.
- First posted on arXiv 27 Nov 2024.

## Builds on (most cited first, 30 of 45)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [FlowNet: Learning Optical Flow with Convolutional Networks](https://mlsanity.com/paper/1504.06852/) (2015)
- [PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume](https://mlsanity.com/paper/1709.02371/) (2017)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks](https://mlsanity.com/paper/1612.01925/) (2016)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR](https://mlsanity.com/paper/2201.12329/) (2022)
- [CoTracker: It is Better to Track Together](https://mlsanity.com/paper/2307.07635/) (2023)
- [DN-DETR: Accelerate DETR Training by Introducing Query DeNoising](https://mlsanity.com/paper/2203.01305/) (2022)
- [Kubric: A scalable dataset generator](https://mlsanity.com/paper/2203.03570/) (2022)
- [TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinement](https://mlsanity.com/paper/2306.08637/) (2023)
- [PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking](https://mlsanity.com/paper/2307.15055/) (2023)
- [FlowFormer: A Transformer Architecture for Optical Flow](https://mlsanity.com/paper/2203.16194/) (2022)
- [TAP-Vid: A Benchmark for Tracking Any Point in a Video](https://mlsanity.com/paper/2211.03726/) (2022)
- [Particle Video Revisited: Tracking Through Occlusions Using Point Trajectories](https://mlsanity.com/paper/2204.04153/) (2022)
- [Tracking Everything Everywhere All at Once](https://mlsanity.com/paper/2306.05422/) (2023)
- [SpatialTracker: Tracking Any 2D Pixels in 3D Space](https://mlsanity.com/paper/2404.04319/) (2024)
- [BootsTAP: Bootstrapped Training for Tracking-Any-Point](https://mlsanity.com/paper/2402.00847/) (2024)
- [The Surprising Effectiveness of Diffusion Models for Optical Flow and Monocular Depth Estimation](https://mlsanity.com/paper/2306.01923/) (2023)
- [FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimation](https://mlsanity.com/paper/2303.01237/) (2023)
- [LLaVA-Grounding: Grounded Visual Chat with Large Multimodal Models](https://mlsanity.com/paper/2312.02949/) (2023)
- [T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy](https://mlsanity.com/paper/2403.14610/) (2024)

## Built on by (most cited first, 8 of 8)

- [TAPNext: Tracking Any Point (TAP) as Next Token Prediction](https://mlsanity.com/paper/2504.05579/) (2025)
- [Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels](https://mlsanity.com/paper/2603.02573/) (2026)
- [TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels](https://mlsanity.com/paper/2512.08358/) (2025)
- [TAPNext++: What's Next for Tracking Any Point (TAP)?](https://mlsanity.com/paper/2604.10582/) (2026)
- [Video Understanding: From Geometry and Semantics to Unified Models](https://mlsanity.com/paper/2603.17840/) (2026)
- [AnthroTAP: Learning Point Tracking with Real-World Motion](https://mlsanity.com/paper/2507.06233/) (2025)
- [Track-On2: Enhancing Online Point Tracking with Memory](https://mlsanity.com/paper/2509.19115/) (2025)
- [TAPVid-MV: A Benchmark for Tracking Any Point in 3D Across Multiple Views](https://mlsanity.com/paper/2609.01899/) (2026)

## Abstract (arXiv)

In this paper, built upon TAPTRv2, we present TAPTRv3. TAPTRv2 is a simple yet effective DETR-like point tracking framework that works fine in regular videos but tends to fail in long videos. TAPTRv3 improves TAPTRv2 by addressing its shortcomings in querying high-quality features from long videos, where the target tracking points normally undergo increasing variation over time. In TAPTRv3, we propose to utilize both spatial and temporal context to bring better feature querying along the spatial and temporal dimensions for more robust tracking in long videos. For better spatial feature querying, we identify that off-the-shelf attention mechanisms struggle with point-level tasks and present Context-aware Cross-Attention (CCA). CCA introduces spatial context into the attention mechanism to enhance the quality of attention scores when querying image features. For better temporal feature querying, we introduce Visibility-aware Long-Temporal Attention (VLTA), which conducts temporal attention over past frames while considering their corresponding visibilities. This effectively addresses the feature drifting problem in TAPTRv2 caused by its RNN-like long-term modeling. TAPTRv3 surpasses TAPTRv2 by a large margin on most of the challenging datasets and obtains state-of-the-art performance. Even when compared with methods trained on large-scale extra internal data, TAPTRv3 still demonstrates superiority.