Delta Sampling R-BERT for limited data and low-light action recognition
arXiv:2107.05202
Abstract
We present an approach to perform supervised action recognition in the dark. In this work, we present our results on the ARID dataset. Most previous works only evaluate performance on large, well illuminated datasets like Kinetics and HMDB51. We demonstrate that our work is able to achieve a very low error rate while being trained on a much smaller dataset of dark videos. We also explore a variety of training and inference strategies including domain transfer methodologies and also propose a simple but useful frame selection strategy. Our empirical results demonstrate that we beat previously published baseline models by 11%.
References in corpus (11)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
- The Kinetics Human Action Video Dataset
- Is Space-Time Attention All You Need for Video Understanding?
- Deep Bilateral Learning for Real-Time Image Enhancement
- A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay
- On the Variance of the Adaptive Learning Rate and Beyond
- Lookahead Optimizer: k steps forward, 1 step back
- Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding
- Gradient Centralization: A New Optimization Technique for Deep Neural Networks
- ARID: A New Dataset for Recognizing Action in the Dark