# Side4Video: Spatial-Temporal Side Network for Memory-Efficient Image-to-Video Transfer Learning

arXiv: https://arxiv.org/abs/2311.15769 · HTML: https://mlsanity.com/paper/2311.15769/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 53 of its 59 references that are in the map.
- First posted on arXiv 27 Nov 2023.

## Builds on (most cited first, 30 of 53)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [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)
- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [Video Swin Transformer](https://mlsanity.com/paper/2106.13230/) (2021)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips](https://mlsanity.com/paper/1906.03327/) (2019)
- [TSM: Temporal Shift Module for Efficient Video Understanding](https://mlsanity.com/paper/1811.08383/) (2018)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [Scaling Vision Transformers to 22 Billion Parameters](https://mlsanity.com/paper/2302.05442/) (2023)
- [CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval](https://mlsanity.com/paper/2104.08860/) (2021)
- [FILIP: Fine-grained Interactive Language-Image Pre-Training](https://mlsanity.com/paper/2111.07783/) (2021)
- [VATEX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language Research](https://mlsanity.com/paper/1904.03493/) (2019)
- [Multi-modal Transformer for Video Retrieval](https://mlsanity.com/paper/2007.10639/) (2020)
- [Expanding Language-Image Pretrained Models for General Video Recognition](https://mlsanity.com/paper/2208.02816/) (2022)
- [Prompting Visual-Language Models for Efficient Video Understanding](https://mlsanity.com/paper/2112.04478/) (2021)

## Built on by (most cited first, 11 of 11)

- [Parameter-Efficient Fine-Tuning for Foundation Models](https://mlsanity.com/paper/2501.13787/) (2025)
- [RAP: Efficient Text-Video Retrieval with Sparse-and-Correlated Adapter](https://mlsanity.com/paper/2405.19465/) (2024)
- [Narrating the Video: Boosting Text-Video Retrieval via Comprehensive Utilization of Frame-Level Captions](https://mlsanity.com/paper/2503.05186/) (2025)
- [Cross-Modal Transfer from Memes to Videos: Addressing Data Scarcity in Hateful Video Detection](https://mlsanity.com/paper/2501.15438/) (2025)
- [DANTE-AD: Dual-Vision Attention Network for Long-Term Audio Description](https://mlsanity.com/paper/2503.24096/) (2025)
- [Feature Hallucination for Self-supervised Action Recognition](https://mlsanity.com/paper/2506.20342/) (2025)
- [A Survey on Backbones for Deep Video Action Recognition](https://mlsanity.com/paper/2405.05584/) (2024)
- [DEAR: Depth-Enhanced Action Recognition](https://mlsanity.com/paper/2408.15679/) (2024)
- [Exploring High-Order Self-Similarity for Video Understanding](https://mlsanity.com/paper/2604.20760/) (2026)
- [TDS-CLIP: Temporal Difference Side Network for Efficient VideoAction Recognition](https://mlsanity.com/paper/2408.10688/) (2024)
- [Human-AI Divergence in Ego-centric Action Recognition under Spatial and Spatiotemporal Manipulations](https://mlsanity.com/paper/2603.08317/) (2026)

## Abstract (arXiv)

Large pre-trained vision models achieve impressive success in computer vision. However, fully fine-tuning large models for downstream tasks, particularly in video understanding, can be prohibitively computationally expensive. Recent studies turn their focus towards efficient image-to-video transfer learning. Nevertheless, existing efficient fine-tuning methods lack attention to training memory usage and exploration of transferring a larger model to the video domain. In this paper, we present a novel Spatial-Temporal Side Network for memory-efficient fine-tuning large image models to video understanding, named Side4Video. Specifically, we introduce a lightweight spatial-temporal side network attached to the frozen vision model, which avoids the backpropagation through the heavy pre-trained model and utilizes multi-level spatial features from the original image model. Extremely memory-efficient architecture enables our method to reduce 75% memory usage than previous adapter-based methods. In this way, we can transfer a huge ViT-E (4.4B) for video understanding tasks which is 14x larger than ViT-L (304M). Our approach achieves remarkable performance on various video datasets across unimodal and cross-modal tasks (i.e., action recognition and text-video retrieval), especially in Something-Something V1&V2 (67.3% & 74.6%), Kinetics-400 (88.6%), MSR-VTT (52.3%), MSVD (56.1%) and VATEX (68.8%). We release our code at https://github.com/HJYao00/Side4Video.