# Learnable Pooling Methods for Video Classification

arXiv: https://arxiv.org/abs/1810.00530 · HTML: https://mlsanity.com/paper/1810.00530/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Oct 2018.

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

- [VLAD-BuFF: Burst-aware Fast Feature Aggregation for Visual Place Recognition](https://mlsanity.com/paper/2409.19293/) (2024)
- [Enhancing Transformer for Video Understanding Using Gated Multi-Level Attention and Temporal Adversarial Training](https://mlsanity.com/paper/2103.10043/) (2021)
- [Leveraging Local Temporal Information for Multimodal Scene Classification](https://mlsanity.com/paper/2110.13992/) (2021)

## Abstract (arXiv)

We introduce modifications to state-of-the-art approaches to aggregating local video descriptors by using attention mechanisms and function approximations. Rather than using ensembles of existing architectures, we provide an insight on creating new architectures. We demonstrate our solutions in the "The 2nd YouTube-8M Video Understanding Challenge", by using frame-level video and audio descriptors. We obtain testing accuracy similar to the state of the art, while meeting budget constraints, and touch upon strategies to improve the state of the art. Model implementations are available in https://github.com/pomonam/LearnablePoolingMethods.