# Learning from One Continuous Video Stream

arXiv: https://arxiv.org/abs/2312.00598 · HTML: https://mlsanity.com/paper/2312.00598/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 49 references that are in the map.
- First posted on arXiv 1 Dec 2023.

## Builds on (most cited first, 30 of 38)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [The 2017 DAVIS Challenge on Video Object Segmentation](https://mlsanity.com/paper/1704.00675/) (2017)
- [Unsupervised Learning of Video Representations using LSTMs](https://mlsanity.com/paper/1502.04681/) (2015)
- [Masked Autoencoders As Spatiotemporal Learners](https://mlsanity.com/paper/2205.09113/) (2022)
- [CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completion](https://mlsanity.com/paper/2210.10716/) (2022)
- [VideoMoCo: Contrastive Video Representation Learning with Temporally Adversarial Examples](https://mlsanity.com/paper/2103.05905/) (2021)
- [An Empirical Investigation of the Role of Pre-training in Lifelong Learning](https://mlsanity.com/paper/2112.09153/) (2021)
- [Revisiting Small Batch Training for Deep Neural Networks](https://mlsanity.com/paper/1804.07612/) (2018)
- [The 2019 DAVIS Challenge on VOS: Unsupervised Multi-Object Segmentation](https://mlsanity.com/paper/1905.00737/) (2019)
- [The CLEAR Benchmark: Continual LEArning on Real-World Imagery](https://mlsanity.com/paper/2201.06289/) (2022)
- [A Short Note on the Kinetics-700-2020 Human Action Dataset](https://mlsanity.com/paper/2010.10864/) (2020)
- [A critical analysis of self-supervision, or what we can learn from a single image](https://mlsanity.com/paper/1904.13132/) (2019)
- [Online Continual Learning with Natural Distribution Shifts: An Empirical Study with Visual Data](https://mlsanity.com/paper/2108.09020/) (2021)
- [Online Prototype Learning for Online Continual Learning](https://mlsanity.com/paper/2308.00301/) (2023)
- [Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning](https://mlsanity.com/paper/2212.03229/) (2022)
- [Audiovisual Masked Autoencoders](https://mlsanity.com/paper/2212.05922/) (2022)
- [Is ImageNet worth 1 video? Learning strong image encoders from 1 long unlabelled video](https://mlsanity.com/paper/2310.08584/) (2023)
- [Test-Time Training on Video Streams](https://mlsanity.com/paper/2307.05014/) (2023)

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

- [Recurrent Video Masked Autoencoders](https://mlsanity.com/paper/2512.13684/) (2025)
- [Learning from Streaming Video with Orthogonal Gradients](https://mlsanity.com/paper/2504.01961/) (2025)
- [Memory Storyboard: Leveraging Temporal Segmentation for Streaming Self-Supervised Learning from Egocentric Videos](https://mlsanity.com/paper/2501.12254/) (2025)
- [MoSiC: Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning](https://mlsanity.com/paper/2506.08694/) (2025)
- [Lifelong Learning of Video Diffusion Models From a Single Video Stream](https://mlsanity.com/paper/2406.04814/) (2024)
- [Video, How Do Your Tokens Merge?](https://mlsanity.com/paper/2506.03885/) (2025)
- [Unique Lives, Shared World: Learning from Single-Life Videos](https://mlsanity.com/paper/2512.04085/) (2025)
- [Human Motion Prediction via Test-domain-aware Adaptation with Easily-available Human Motions Estimated from Videos](https://mlsanity.com/paper/2505.07301/) (2025)
- [Sequence-Adaptive Video Prediction in Continuous Streams using Diffusion Noise Optimization](https://mlsanity.com/paper/2511.18255/) (2025)
- [FIVE-VLA: Fast and EffectIVE Autonomous Driving with Recurrent Action Memory](https://mlsanity.com/paper/2609.18623/) (2026)
- [I Have a Stream: Making Self-Supervised Learning Work on Continuous Video](https://mlsanity.com/paper/2609.40333/) (2026)

## Abstract (arXiv)

We introduce a framework for online learning from a single continuous video stream -- the way people and animals learn, without mini-batches, data augmentation or shuffling. This poses great challenges given the high correlation between consecutive video frames and there is very little prior work on it. Our framework allows us to do a first deep dive into the topic and includes a collection of streams and tasks composed from two existing video datasets, plus methodology for performance evaluation that considers both adaptation and generalization. We employ pixel-to-pixel modelling as a practical and flexible way to switch between pre-training and single-stream evaluation as well as between arbitrary tasks, without ever requiring changes to models and always using the same pixel loss. Equipped with this framework we obtained large single-stream learning gains from pre-training with a novel family of future prediction tasks, found that momentum hurts, and that the pace of weight updates matters. The combination of these insights leads to matching the performance of IID learning with batch size 1, when using the same architecture and without costly replay buffers.