# Learned Image Compression for Machine Perception

arXiv: https://arxiv.org/abs/2111.02249 · HTML: https://mlsanity.com/paper/2111.02249/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 49 references that are in the map.
- First posted on arXiv 3 Nov 2021.

## Builds on (most cited first, 26 of 26)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network](https://mlsanity.com/paper/1609.05158/) (2016)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learning](https://mlsanity.com/paper/1702.03118/) (2017)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Joint Autoregressive and Hierarchical Priors for Learned Image Compression](https://mlsanity.com/paper/1809.02736/) (2018)
- [BinaryConnect: Training Deep Neural Networks with binary weights during propagations](https://mlsanity.com/paper/1511.00363/) (2015)
- [Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules](https://mlsanity.com/paper/2001.01568/) (2020)
- [High-Fidelity Generative Image Compression](https://mlsanity.com/paper/2006.09965/) (2020)
- [Frustratingly Simple Few-Shot Object Detection](https://mlsanity.com/paper/2003.06957/) (2020)
- [Generative Adversarial Networks for Extreme Learned Image Compression](https://mlsanity.com/paper/1804.02958/) (2018)
- [Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoff](https://mlsanity.com/paper/1901.07821/) (2019)
- [Meta R-CNN : Towards General Solver for Instance-level Few-shot Learning](https://mlsanity.com/paper/1909.13032/) (2019)
- [Full Resolution Image Compression with Recurrent Neural Networks](https://mlsanity.com/paper/1608.05148/) (2016)
- [Scalable Methods for 8-bit Training of Neural Networks](https://mlsanity.com/paper/1805.11046/) (2018)
- [Density Modeling of Images using a Generalized Normalization Transformation](https://mlsanity.com/paper/1511.06281/) (2015)
- [Practical Full Resolution Learned Lossless Image Compression](https://mlsanity.com/paper/1811.12817/) (2018)
- [Improved Lossy Image Compression with Priming and Spatially Adaptive Bit Rates for Recurrent Networks](https://mlsanity.com/paper/1703.10114/) (2017)
- [Towards Image Understanding from Deep Compression without Decoding](https://mlsanity.com/paper/1803.06131/) (2018)
- [End-to-end Learning of Compressible Features](https://mlsanity.com/paper/2007.11797/) (2020)
- [Spatially adaptive image compression using a tiled deep network](https://mlsanity.com/paper/1802.02629/) (2018)

## Built on by (most cited first, 14 of 14)

- [Image Coding for Machines with Omnipotent Feature Learning](https://mlsanity.com/paper/2207.01932/) (2022)
- [Image Compression for Machine and Human Vision with Spatial-Frequency Adaptation](https://mlsanity.com/paper/2407.09853/) (2024)
- [Image Coding for Machines with Edge Information Learning Using Segment Anything](https://mlsanity.com/paper/2403.04173/) (2024)
- [Perceptual Video Coding for Machines via Satisfied Machine Ratio Modeling](https://mlsanity.com/paper/2211.06797/) (2022)
- [Image Coding for Machines with Object Region Learning](https://mlsanity.com/paper/2308.13984/) (2023)
- [NeuCo-Bench: A Novel Benchmark Framework for Neural Embeddings in Earth Observation](https://mlsanity.com/paper/2510.17914/) (2025)
- [Machine Perception-Driven Image Compression: A Layered Generative Approach](https://mlsanity.com/paper/2304.06896/) (2023)
- [Scalable Image Coding for Humans and Machines Using Feature Fusion Network](https://mlsanity.com/paper/2405.09152/) (2024)
- [Delta-ICM: Entropy Modeling with Delta Function for Learned Image Compression](https://mlsanity.com/paper/2410.07669/) (2024)
- [Multiscale Feature Importance-based Bit Allocation for End-to-End Feature Coding for Machines](https://mlsanity.com/paper/2503.19278/) (2025)
- [Guided Diffusion for the Extension of Machine Vision to Human Visual Perception](https://mlsanity.com/paper/2503.17907/) (2025)
- [Learned Image Compression with Text Quality Enhancement](https://mlsanity.com/paper/2402.08643/) (2024)
- [Deep-Learned Compression for Radio-Frequency Signal Classification](https://mlsanity.com/paper/2403.03150/) (2024)
- [What and Where to Adapt: Structure-Semantics Co-Tuning for Machine Vision Compression via Synergistic Adapters](https://mlsanity.com/paper/2604.10017/) (2026)

## Abstract (arXiv)

Recent work has shown that learned image compression strategies can outperform standard hand-crafted compression algorithms that have been developed over decades of intensive research on the rate-distortion trade-off. With growing applications of computer vision, high quality image reconstruction from a compressible representation is often a secondary objective. Compression that ensures high accuracy on computer vision tasks such as image segmentation, classification, and detection therefore has the potential for significant impact across a wide variety of settings. In this work, we develop a framework that produces a compression format suitable for both human perception and machine perception. We show that representations can be learned that simultaneously optimize for compression and performance on core vision tasks. Our approach allows models to be trained directly from compressed representations, and this approach yields increased performance on new tasks and in low-shot learning settings. We present results that improve upon segmentation and detection performance compared to standard high quality JPGs, but with representations that are four to ten times smaller in terms of bits per pixel. Further, unlike naive compression methods, at a level ten times smaller than standard JEPGs, segmentation and detection models trained from our format suffer only minor degradation in performance.