mlsanityData to 7 Oct 2026

Deep Contextual Video Compression

arXiv 2109.15047 ↗First posted 30 Sep 2021eess.IV cs.CV cs.MMcited by 216

Not on the map. Why?

Abstract (arXiv)

Most of the existing neural video compression methods adopt the predictive coding framework, which first generates the predicted frame and then encodes its residue with the current frame. However, as for compression ratio, predictive coding is only a sub-optimal solution as it uses simple subtraction operation to remove the redundancy across frames. In this paper, we propose a deep contextual video compression framework to enable a paradigm shift from predictive coding to conditional coding. In particular, we try to answer the following questions: how to define, use, and learn condition under a deep video compression framework. To tap the potential of conditional coding, we propose using feature domain context as condition. This enables us to leverage the high dimension context to carry rich information to both the encoder and the decoder, which helps reconstruct the high-frequency contents for higher video quality. Our framework is also extensible, in which the condition can be flexibly designed. Experiments show that our method can significantly outperform the previous state-of-the-art (SOTA) deep video compression methods. When compared with x265 using veryslow preset, we can achieve 26.0% bitrate saving for 1080P standard test videos.

Key facts. Cited by 216 papers in the map (data through 7 Oct 2026). Builds on 26 of its 39 references that are in the map. First posted on arXiv 30 Sep 2021.

Builds on

26 of 39 references.

References · 26

  1. 2021Transformers in Vision: A Surveycited by 1,009
  2. 2018Variational image compression with a scale hyperpriorcited by 910
  3. 2016End-to-end Optimized Image Compressioncited by 673
  4. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 642
  5. 2017Video Enhancement with Task-Oriented Flowcited by 612
  6. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 570
  7. 2016Optical Flow Estimation using a Spatial Pyramid Networkcited by 421
  8. 2018DVC: An End-to-end Deep Video Compression Frameworkcited by 324
  9. 2017Lossy Image Compression with Compressive Autoencoderscited by 289
  10. 2020Softmax Splatting for Video Frame Interpolationcited by 276
  11. 2020CompressAI: a PyTorch library and evaluation platform for end-to-end compression researchcited by 271
  12. 2019Depth-Aware Video Frame Interpolationcited by 232
14 more
  1. 2016Full Resolution Image Compression with Recurrent Neural Networkscited by 170
  2. 2015Variable Rate Image Compression with Recurrent Neural Networkscited by 168
  3. 2018Video Compression through Image Interpolationcited by 143
  4. 2020M-LVC: Multiple Frames Prediction for Learned Video Compressioncited by 106
  5. 2019Video Compression With Rate-Distortion Autoencoderscited by 97
  6. 2017Improved Lossy Image Compression with Priming and Spatially Adaptive Bit Rates for Recurrent Networkscited by 91
  7. 2020Learning for Video Compression with Hierarchical Quality and Recurrent Enhancementcited by 90
  8. 2018Learned Video Compressioncited by 89
  9. 2020Content Adaptive and Error Propagation Aware Deep Video Compressioncited by 72
  10. 2020Improving Deep Video Compression by Resolution-adaptive Flow Codingcited by 58
  11. 2020Conditional Entropy Coding for Efficient Video Compressioncited by 42
  12. 2020Hierarchical Autoregressive Modeling for Neural Video Compressioncited by 28
  13. 2021Conditional Coding for Flexible Learned Video Compressioncited by 25
  14. 2020Optical Flow and Mode Selection for Learning-based Video Codingcited by 15

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 26; 2023: 35; 2024: 52; 2025: 53; 2026: 47.

216 papers.

Sort:

By area:

  1. 2024HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compressioncited by 202
  2. 2023Neural Video Compression with Diverse Contextscited by 155
  3. 2022Hybrid Spatial-Temporal Entropy Modelling for Neural Video Compressioncited by 123
  4. 2024Neural Video Compression with Feature Modulationcited by 115
  5. 2021Temporal Context Mining for Learned Video Compressioncited by 112
  6. 2024Image and Video Tokenization with Binary Spherical Quantizationcited by 110
  7. 2023HNeRV: A Hybrid Neural Representation for Videoscited by 100
  8. 2024CompGS: Efficient 3D Scene Representation via Compressed Gaussian Splattingcited by 80
  9. 2022VCT: A Video Compression Transformercited by 74
  10. 2025Improving the Diffusability of Autoencoderscited by 74
  11. 2021Nonlinear Transform Source-Channel Coding for Semantic Communicationscited by 73
  12. 2025Towards Practical Real-Time Neural Video Compressioncited by 72
38 more
  1. 2023HiNeRV: Video Compression with Hierarchical Encoding-based Neural Representationcited by 70
  2. 2022Coarse-to-fine Deep Video Coding with Hyperprior-guided Mode Predictioncited by 55
  3. 2023C3: High-performance and low-complexity neural compression from a single image or videocited by 54
  4. 2022Wireless Deep Video Semantic Transmissioncited by 49
  5. 2022CANF-VC: Conditional Augmented Normalizing Flows for Video Compressioncited by 44
  6. 2024ElasticTok: Adaptive Tokenization for Image and Videocited by 44
  7. 2023DNeRV: Modeling Inherent Dynamics via Difference Neural Representation for Videoscited by 42
  8. 2023High-Fidelity and Freely Controllable Talking Head Video Generationcited by 42
  9. 2023Towards Scalable Neural Representation for Diverse Videoscited by 37
  10. 2024USTC-TD: A Test Dataset and Benchmark for Image and Video Coding in 2020scited by 37
  11. 2022PS-NeRV: Patch-wise Stylized Neural Representations for Videoscited by 32
  12. 2024HEMGS: A Hybrid Entropy Model for 3D Gaussian Splatting Data Compressioncited by 31
  13. 2024How Far Can We Compress Instant-NGP-Based NeRF?cited by 30
  14. 2024Boosting Neural Representations for Videos with a Conditional Decodercited by 29
  15. 2024NVRC: Neural Video Representation Compressioncited by 29
  16. 2022AlphaVC: High-Performance and Efficient Learned Video Compressioncited by 28
  17. 2024DS-NeRV: Implicit Neural Video Representation with Decomposed Static and Dynamic Codescited by 25
  18. 2022Neural Compression-Based Feature Learning for Video Restorationcited by 23
  19. 2022CaDM: Codec-aware Diffusion Modeling for Neural-enhanced Video Streamingcited by 23
  20. 2025GIFStream: 4D Gaussian-based Immersive Video with Feature Streamcited by 23
  21. 2025Generative Latent Coding for Ultra-Low Bitrate Image and Video Compressioncited by 22
  22. 2024Spatial Decomposition and Temporal Fusion based Inter Prediction for Learned Video Compressioncited by 21
  23. 2025Generative Latent Coding for Ultra-Low Bitrate Image Compressioncited by 21
  24. 2024Lossless data compression by large modelscited by 20
  25. 2024Bi-Directional Deep Contextual Video Compressioncited by 20
  26. 2025Learned Image Compression with Hierarchical Progressive Context Modelingcited by 20
  27. 2024PNVC: Towards Practical INR-based Video Compressioncited by 19
  28. 2022B-CANF: Adaptive B-frame Coding with Conditional Augmented Normalizing Flowscited by 18
  29. 2022A Coding Framework and Benchmark towards Low-Bitrate Video Understandingcited by 17
  30. 2023Offline and Online Optical Flow Enhancement for Deep Video Compressioncited by 17
  31. 2025Neural Video Compression with Context Modulationcited by 15
  32. 2022Advancing Learned Video Compression with In-loop Frame Predictioncited by 14
  33. 2023Hierarchical B-frame Video Coding Using Two-Layer CANF without Motion Codingcited by 14
  34. 2023Multi-Modality Deep Network for Extreme Learned Image Compressioncited by 14
  35. 2022Task-Oriented Communication for Edge Video Analyticscited by 13
  36. 2023VNVC: A Versatile Neural Video Coding Framework for Efficient Human-Machine Visioncited by 13
  37. 2021Learning Cross-Scale Weighted Prediction for Efficient Neural Video Compressioncited by 12
  38. 2022Neighbor Correspondence Matching for Flow-based Video Frame Synthesiscited by 12

50 most cited of 216 All 216 (JSON)