mlsanityData to 7 Oct 2026

Extreme Compression of Adaptive Neural Images

arXiv 2405.16807 ↗First posted 27 May 2024cs.CV cs.AI cs.GRcited by 1

Not on the map. Why?

Abstract (arXiv)

Implicit Neural Representations (INRs) and Neural Fields are a novel paradigm for signal representation, from images and audio to 3D scenes and videos. The fundamental idea is to represent a signal as a continuous and differentiable neural network. This new approach poses new theoretical questions and challenges. Considering a neural image as a 2D image represented as a neural network, we aim to explore novel neural image compression. In this work, we present a novel analysis on compressing neural fields, with focus on images and introduce Adaptive Neural Images (ANI), an efficient neural representation that enables adaptation to different inference or transmission requirements. Our proposed method allows us to reduce the bits-per-pixel (bpp) of the neural image by 8 times, without losing sensitive details or harming fidelity. Our work offers a new framework for developing compressed neural fields. We achieve a new state-of-the-art in terms of PSNR/bpp trade-off thanks to our successful implementation of 4-bit neural representations.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 50 of its 62 references that are in the map. First posted on arXiv 27 May 2024.

Builds on

50 of 62 references.

References · 50

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
  3. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,355
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  5. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  7. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  8. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  9. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  10. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  11. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,338
  12. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
38 more
  1. 2018Variational image compression with a scale hyperpriorcited by 910
  2. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  3. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
  4. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  5. 2016End-to-end Optimized Image Compressioncited by 673
  6. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 642
  7. 2020Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humanscited by 634
  8. 2019Learned Step Size Quantizationcited by 604
  9. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 600
  10. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 426
  11. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 387
  12. 2020High-Fidelity Generative Image Compressioncited by 331
  13. 2021NeRV: Neural Representations for Videoscited by 255
  14. 2019Learning Shape Templates with Structured Implicit Functionscited by 242
  15. 2020Learned Initializations for Optimizing Coordinate-Based Neural Representationscited by 240
  16. 2021COIN: COmpression with Implicit Neural representationscited by 232
  17. 2023WIRE: Wavelet Implicit Neural Representationscited by 220
  18. 2018Generative Adversarial Networks for Extreme Learned Image Compressioncited by 213
  19. 2020BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Modelscited by 187
  20. 2019Low-bit Quantization of Neural Networks for Efficient Inferencecited by 185
  21. 2020LSQ+: Improving low-bit quantization through learnable offsets and better initializationcited by 180
  22. 2021Implicit Neural Representations for Image Compressioncited by 161
  23. 2021Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Bettercited by 153
  24. 2018Context-adaptive Entropy Model for End-to-end Optimized Image Compressioncited by 138
  25. 2018Conditional Probability Models for Deep Image Compressioncited by 128
  26. 2022VideoINR: Learning Video Implicit Neural Representation for Continuous Space-Time Super-Resolutioncited by 101
  27. 2019Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networkscited by 87
  28. 2022COIN++: Neural Compression Across Modalitiescited by 81
  29. 2022Transformers as Meta-Learners for Implicit Neural Representationscited by 79
  30. 2022Multi-Realism Image Compression with a Conditional Generatorcited by 79
  31. 2023High-Fidelity Image Compression with Score-based Generative Modelscited by 53
  32. 2021Meta-Learning Sparse Implicit Neural Representationscited by 48
  33. 2023SHACIRA: Scalable HAsh-grid Compression for Implicit Neural Representationscited by 31
  34. 2022DINER: Disorder-Invariant Implicit Neural Representationcited by 30
  35. 2022On Quantizing Implicit Neural Representationscited by 20
  36. 2022HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworkscited by 12
  37. 2021Quantization-Guided Training for Compact TinyML Modelscited by 10
  38. 2020Learning to Improve Image Compression without Changing the Standard Decodercited by 7

Built on by

Cited by year

Citing papers per year: 2024: 1.

1 paper.

  1. 2024Streaming Neural Imagescited by 0