AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Exploring Simple and Transferable Recognition-Aware Image Processing

Zhuang Liu, Hung-Ju Wang, Tinghui Zhou, Zhiqiang Shen, Bingyi Kang, Evan Shelhamer, Trevor Darrell

TPAMI 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Human-like visual cognition in VLMs · 27 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
add recognition loss from a fixed pretrained recognizerSRResNet super-resolution · ImageNet val (ResNet-18)top-1 accuracy: 52.6 -> 61.8Improved
add recognition loss, JPEG-deblockingSRResNet · ImageNet val (ResNet-18)top-1 accuracy: 48.2 -> 57.7Improved
add recognition loss, denoisingSRResNet · ImageNet val (ResNet-18)top-1 accuracy: 61.9 -> 65.1Improved
add an intermediate transformer between processor and recognizerRA processing, super-resolution · ImageNet val (ResNet-18)top-1 accuracy: 61.8 -> 63.0Improved
use unsupervised RA (distill recognizer output) instead of label-based RARA processing, super-resolution · ImageNet val (ResNet-18)top-1 accuracy: 61.8 -> 61.3Hurt
train the processor with recognizer ResNet-18 and evaluate on ResNet-50RA processing, super-resolution · ImageNet val (ResNet-50)top-1 accuracy: 58.8 -> 66.7Improved
transformer variant trained on ResNet-18, evaluated on VGG-16RA w/ Transformer, super-resolution · ImageNet val (VGG-16)top-1 accuracy: 50.2 -> 27.0Hurt
add RA loss when transferring across disjoint category splitsRA processing, super-resolution · ImageNet Cat A to Cat B500-way accuracy: 60.1 -> 66.5Improved
add RA loss on randomized compound corruptionRA processing · ImageNet val (ResNet-18)top-1 accuracy: 31.5 -> 41.2Improved
set lambda to 1e-3 instead of 0 for image quality tradeoff (ablation)RA processing, super-resolution · ImageNet SR PSNRPSNR: 26.73 -> 26.31Hurt
raise lambda from 1e-3 to 1e-2 (ablation)RA processing, super-resolution · ImageNet val (ResNet-18)top-1 accuracy: 61.8 -> 61.3Hurt
add the image processing loss alongside the recognition loss (ablation)RA processing, super-resolution · ImageNet val (ResNet-18)top-1 accuracy: 60.9 -> 61.8Improved

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