AI Paper Atlas

Research map / Architectures and efficiency / Quantization

QARV: Quantization-Aware ResNet VAE for Lossy Image Compression

Zhihao Duan, Ming Lu, Jack Ma, Yuning Huang, Zhan Ma, Fengqing Maggie Zhu

TPAMI 2023 · Architectures and efficiency › Quantization · approach: Post-training weight quantization · 107 citations (OpenAlex)

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 changedWhereMeasuredEffect
place affine transform after layer norm (position 2, AdaLN) instead of position 0 (ablation)QARV · KodakBD-rate (%): 2.446 -> -0.066Improved
AdaLN position 2 vs position 1 (ablation)QARV · KodakBD-rate (%): 2.005 -> -0.066Improved
AdaLN position 2 vs position 4 (ablation)QARV · KodakBD-rate (%): 5.365 -> -0.066Improved
use Layer Norm instead of Group Norm (ablation)QARV · KodakBD-rate (%): 1.041 -> -0.066Improved
use Layer Norm instead of Instance Norm (ablation)QARV · KodakBD-rate (%): 10.26 -> -0.066Improved
add bi-directional inference to latent block (config B vs A), N=2 (ablation)QARV · Kodak (N=2)BD-rate (%): 24.96 -> 18.20Improved
add residual connection to latent block (config C vs B), N=2 (ablation)QARV · Kodak (N=2)BD-rate (%): 18.20 -> 14.06Improved
add bi-directional inference (B vs A), N=3 (ablation)QARV · Kodak (N=3)BD-rate (%): 22.51 -> 9.076Improved
add residual connection (C vs B), N=3 (ablation)QARV · Kodak (N=3)BD-rate (%): 9.076 -> 3.761Improved
increase number of latent variables from N=2 to N=3 with config A (ablation)Hyperprior-style config A · KodakBD-rate (%): 24.96 -> 22.51Improved
use ConvNeXt instead of NAT residual blocks (ablation)QARV · KodakBD-rate (%): 2.916 -> -0.066Improved
sample lambda by 1/3-power uniform instead of log-uniform (ablation)QARV · KodakBD-rate (%): -4.28 -> -4.74Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy