# Exploring Stronger Feature for Temporal Action Localization

arXiv: https://arxiv.org/abs/2106.13014 · HTML: https://mlsanity.com/paper/2106.13014/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 27 references that are in the map.
- First posted on arXiv 24 Jun 2021.

## Builds on (most cited first, 23 of 23)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [Soft-NMS -- Improving Object Detection With One Line of Code](https://mlsanity.com/paper/1704.04503/) (2017)
- [A Short Note about Kinetics-600](https://mlsanity.com/paper/1808.01340/) (2018)
- [BMN: Boundary-Matching Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/1907.09702/) (2019)
- [A Short Note on the Kinetics-700 Human Action Dataset](https://mlsanity.com/paper/1907.06987/) (2019)
- [Video Classification with Channel-Separated Convolutional Networks](https://mlsanity.com/paper/1904.02811/) (2019)
- [BSN: Boundary Sensitive Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/1806.02964/) (2018)
- [G-TAD: Sub-Graph Localization for Temporal Action Detection](https://mlsanity.com/paper/1911.11462/) (2019)
- [Self-supervised Co-training for Video Representation Learning](https://mlsanity.com/paper/2010.09709/) (2020)
- [TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals](https://mlsanity.com/paper/1703.06189/) (2017)
- [Temporal Context Aggregation Network for Temporal Action Proposal Refinement](https://mlsanity.com/paper/2103.13141/) (2021)
- [Fast Learning of Temporal Action Proposal via Dense Boundary Generator](https://mlsanity.com/paper/1911.04127/) (2019)
- [Boundary Content Graph Neural Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/2008.01432/) (2020)
- [Self-Supervised Learning for Semi-Supervised Temporal Action Proposal](https://mlsanity.com/paper/2104.03214/) (2021)
- [Self-supervised Motion Learning from Static Images](https://mlsanity.com/paper/2104.00240/) (2021)
- [Towards Training Stronger Video Vision Transformers for EPIC-KITCHENS-100 Action Recognition](https://mlsanity.com/paper/2106.05058/) (2021)
- [CBR-Net: Cascade Boundary Refinement Network for Action Detection: Submission to ActivityNet Challenge 2020 (Task 1)](https://mlsanity.com/paper/2006.07526/) (2020)
- [Temporal Fusion Network for Temporal Action Localization:Submission to ActivityNet Challenge 2020 (Task E)](https://mlsanity.com/paper/2006.07520/) (2020)
- [Augmenting Proposals by the Detector Itself](https://mlsanity.com/paper/2101.11789/) (2021)

## Built on by (most cited first, 3 of 3)

- [TAda! Temporally-Adaptive Convolutions for Video Understanding](https://mlsanity.com/paper/2110.06178/) (2021)
- [Vision Transformers for Action Recognition: A Survey](https://mlsanity.com/paper/2209.05700/) (2022)
- [Temporally-Adaptive Models for Efficient Video Understanding](https://mlsanity.com/paper/2308.05787/) (2023)

## Abstract (arXiv)

Temporal action localization aims to localize starting and ending time with action category. Limited by GPU memory, mainstream methods pre-extract features for each video. Therefore, feature quality determines the upper bound of detection performance. In this technical report, we explored classic convolution-based backbones and the recent surge of transformer-based backbones. We found that the transformer-based methods can achieve better classification performance than convolution-based, but they cannot generate accuracy action proposals. In addition, extracting features with larger frame resolution to reduce the loss of spatial information can also effectively improve the performance of temporal action localization. Finally, we achieve 42.42% in terms of mAP on validation set with a single SlowFast feature by a simple combination: BMN+TCANet, which is 1.87% higher than the result of 2020's multi-model ensemble. Finally, we achieve Rank 1st on the CVPR2021 HACS supervised Temporal Action Localization Challenge.