# Seeing in the Dark: A Teacher-Student Framework for Dark Video Action Recognition via Knowledge Distillation and Contrastive Learning

arXiv: https://arxiv.org/abs/2502.03724 · HTML: https://mlsanity.com/paper/2502.03724/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation.
- Builds on 28 of its 47 references that are in the map.
- First posted on arXiv 6 Feb 2025.

## Placed in

- Knowledge distillation / [General knowledge distillation methods](https://mlsanity.com/distill/general-knowledge-distillation-methods/)

## Builds on (most cited first, 28 of 28)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [Two-Stream Convolutional Networks for Action Recognition in Videos](https://mlsanity.com/paper/1406.2199/) (2014)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?](https://mlsanity.com/paper/1711.09577/) (2017)
- [Born Again Neural Networks](https://mlsanity.com/paper/1805.04770/) (2018)
- [Learning Spatio-Temporal Representation with Pseudo-3D Residual Networks](https://mlsanity.com/paper/1711.10305/) (2017)
- [Convolutional Two-Stream Network Fusion for Video Action Recognition](https://mlsanity.com/paper/1604.06573/) (2016)
- [Spatiotemporal Contrastive Video Representation Learning](https://mlsanity.com/paper/2008.03800/) (2020)
- [Variational Information Distillation for Knowledge Transfer](https://mlsanity.com/paper/1904.05835/) (2019)
- [VideoMoCo: Contrastive Video Representation Learning with Temporally Adversarial Examples](https://mlsanity.com/paper/2103.05905/) (2021)
- [Targeted Supervised Contrastive Learning for Long-Tailed Recognition](https://mlsanity.com/paper/2111.13998/) (2021)
- [Large-scale weakly-supervised pre-training for video action recognition](https://mlsanity.com/paper/1905.00561/) (2019)
- [Audio-Visual Scene-Aware Dialog](https://mlsanity.com/paper/1901.09107/) (2019)
- [D3D: Distilled 3D Networks for Video Action Recognition](https://mlsanity.com/paper/1812.08249/) (2018)
- [ARID: A New Dataset for Recognizing Action in the Dark](https://mlsanity.com/paper/2006.03876/) (2020)
- [Learning an Augmented RGB Representation with Cross-Modal Knowledge Distillation for Action Detection](https://mlsanity.com/paper/2108.03619/) (2021)
- [DistInit: Learning Video Representations Without a Single Labeled Video](https://mlsanity.com/paper/1901.09244/) (2019)
- [Delta Sampling R-BERT for limited data and low-light action recognition](https://mlsanity.com/paper/2107.05202/) (2021)
- [OOD Aware Supervised Contrastive Learning](https://mlsanity.com/paper/2310.01942/) (2023)
- [ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos](https://mlsanity.com/paper/2404.06243/) (2024)

## Abstract (arXiv)

Action recognition in dark or low-light (under-exposed) videos is a challenging task due to visibility degradation, which can hinder critical spatiotemporal details. This paper proposes ActLumos, a teacher-student framework that attains single-stream inference while retaining multi-stream level accuracy. The teacher consumes dual stream inputs, which include original dark frames and retinex-enhanced frames, processed by weight-shared R(2+1)D-34 backbones and fused by a Dynamic Feature Fusion (DFF) module, which dynamically re-weights the two streams at each time step, emphasising the most informative temporal segments. The teacher is also included with a supervised contrastive loss (SupCon) that sharpens class margins. The student shares the R(2+1)D-34 backbone but uses only dark frames and no fusion at test time. The student is first pre-trained with self-supervision on dark clips of both datasets without their labels and then fine-tuned with knowledge distillation from the teacher, transferring the teacher's multi-stream knowledge into a single-stream model. Under single-stream inference, the distilled student attains state-of-the-art accuracy of 96.92% (Top-1) on ARID V1.0, 88.27% on ARID V1.5, and 48.96% on Dark48. Ablation studies further highlight the individual contributions of each component, i.e., DFF in the teacher outperforms single or static fusion, knowledge distillation (KD) transfers these gains to the single-stream student, and two-view spatio-temporal SSL surpasses spatial-only or temporal-only variants without increasing inference cost. The official website of this work is available at: https://github.com/HrishavBakulBarua/ActLumos