Research map / Image generation / Image restoration
Task-Aware Quantization Network for JPEG Image Compression
Jinyoung Choi, Bohyung Han
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train the quantization table with a classification (cross-entropy) loss instead of standard JPEG tables | JPEG compression · ImageNet val classification | accuracy at equal bitrate | Improved |
| add a CBAM-like attention module at the end of the quantization CNN (ablation) | Task-aware quantization network · ImageNet val classification | accuracy vs bitrate | Improved |
| interpolate quantization tables from two trained models for bitrate control (ablation) | Task-aware quantization network · ImageNet val classification | accuracy vs bitrate | No clear effect |
| train with Inception-v3 perceptual loss for captioning | Task-aware quantization network · Flickr8k captioning | BLEU | Improved |
| train with MSE or MS-SSIM loss instead of standard JPEG tables | Task-aware quantization network · Kodak and ImageNet val | PSNR and MS-SSIM | Improved |
| apply task-aware quantization at higher bitrates (above about 0.4 bpp) (ablation) | Task-aware quantization network · ImageNet val classification | accuracy gain | No clear effect |
Closest papers by meaning
- JPEG Inspired Deep Learning (ICLR 2025 · similarity 0.89)
- Learning Dual Priors for JPEG Compression Artifacts Removal (ICCV 2021 · similarity 0.88)
- Variable Rate Deep Image Compression With a Conditional Autoencoder (ICCV 2019 · similarity 0.87)
- End-to-end Optimized Image Compression (ICLR 2017 · similarity 0.87)
- Improved Lossy Image Compression With Priming and Spatially Adaptive Bit Rates for Recurrent Networks (CVPR 2018 · similarity 0.87)
- Focused Quantization for Sparse CNNs (NeurIPS 2019 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
