# MeMSVD: Long-Range Temporal Structure Capturing Using Incremental SVD

arXiv: https://arxiv.org/abs/2406.07191 · HTML: https://mlsanity.com/paper/2406.07191/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 11 Jun 2024.

## Builds on (most cited first, 20 of 20)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Multiscale Vision Transformers](https://mlsanity.com/paper/2104.11227/) (2021)
- [TSM: Temporal Shift Module for Efficient Video Understanding](https://mlsanity.com/paper/1811.08383/) (2018)
- [Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding](https://mlsanity.com/paper/1604.01753/) (2016)
- [MViTv2: Improved Multiscale Vision Transformers for Classification and Detection](https://mlsanity.com/paper/2112.01526/) (2021)
- [AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions](https://mlsanity.com/paper/1705.08421/) (2017)
- [Temporal Relational Reasoning in Videos](https://mlsanity.com/paper/1711.08496/) (2017)
- [Video Action Transformer Network](https://mlsanity.com/paper/1812.02707/) (2018)
- [Long-Term Feature Banks for Detailed Video Understanding](https://mlsanity.com/paper/1812.05038/) (2018)
- [MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video Recognition](https://mlsanity.com/paper/2201.08383/) (2022)
- [Towards Long-Form Video Understanding](https://mlsanity.com/paper/2106.11310/) (2021)
- [SOFT: Softmax-free Transformer with Linear Complexity](https://mlsanity.com/paper/2110.11945/) (2021)
- [A Short Note on the Kinetics-700-2020 Human Action Dataset](https://mlsanity.com/paper/2010.10864/) (2020)
- [Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization](https://mlsanity.com/paper/2006.07976/) (2020)
- [Is Attention Better Than Matrix Decomposition?](https://mlsanity.com/paper/2109.04553/) (2021)
- [Asynchronous Interaction Aggregation for Action Detection](https://mlsanity.com/paper/2004.07485/) (2020)
- [Beyond Short Clips: End-to-End Video-Level Learning with Collaborative Memories](https://mlsanity.com/paper/2104.01198/) (2021)
- [Multiscale Vision Transformers meet Bipartite Matching for efficient single-stage Action Localization](https://mlsanity.com/paper/2312.17686/) (2023)

## Abstract (arXiv)

This paper is on long-term video understanding where the goal is to recognise human actions over long temporal windows (up to minutes long). In prior work, long temporal context is captured by constructing a long-term memory bank consisting of past and future video features which are then integrated into standard (short-term) video recognition backbones through the use of attention mechanisms. Two well-known problems related to this approach are the quadratic complexity of the attention operation and the fact that the whole feature bank must be stored in memory for inference. To address both issues, we propose an alternative to attention-based schemes which is based on a low-rank approximation of the memory obtained using Singular Value Decomposition. Our scheme has two advantages: (a) it reduces complexity by more than an order of magnitude, and (b) it is amenable to an efficient implementation for the calculation of the memory bases in an incremental fashion which does not require the storage of the whole feature bank in memory. The proposed scheme matches or surpasses the accuracy achieved by attention-based mechanisms while being memory-efficient. Through extensive experiments, we demonstrate that our framework generalises to different architectures and tasks, outperforming the state-of-the-art in three datasets.