AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Learning Affinity-Aware Upsampling for Deep Image Matting

Yutong Dai, Hao Lu, Chunhua Shen

CVPR 2021 · Architectures and efficiency › CNN design · approach: Dense matching and estimation CNNs · 65 citations (OpenAlex)

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
replace nearest-neighbor upsampling with affinity-aware dynamic-cs-d upsamplingResNet34 matting network · Composition-1k testSAD: 37.51 -> 32.15Improved
replace nearest-neighbor upsampling with paired affinity-aware downsampling and upsamplingResNet34 matting network · Distinctions-646 testSAD: 25.03 -> 23.20Improved
use affinity-aware upsampling instead of IndexNetMaxPool reconstruction network · MNISTPSNR: 45.51 -> 47.63Improved
use affinity-aware upsampling instead of IndexNetMaxPool reconstruction network · Fashion-MNISTPSNR: 45.83 -> 46.41Improved
pair dynamic-cs downsampling with upsampling (ablation)A²U matting network · Composition-1k testSAD: 35.86 -> 33.13Improved
add normalization and nonlinearity after the encoding layers (ablation)A²U dynamic-cs-d matting network · Composition-1k testSAD: 33.13 -> 32.15Improved
use channel-wise rather than channel-shared encoding in the hybrid implementation (ablation)A²U matting network · Composition-1k testSAD: 36.43 -> 34.76Improved
use channel-shared rather than channel-wise encoding in the dynamic implementation (ablation)A²U matting network · Composition-1k testSAD: 36.66 -> 35.86Improved
set the upsampling kernel size to 3 rather than 1 (ablation)A²U hybrid-cw matting network · Composition-1k testSAD: 37.74 -> 34.76Improved
increase the upsampling kernel size from 3 to 5 (ablation)A²U hybrid-cw matting network · Composition-1k testSAD: 34.76 -> 35.99Hurt
use sigmoid+softmax instead of softmax normalization (ablation)A²U hybrid-cw matting network · Composition-1k testSAD: 35.93 -> 34.76Improved
use sigmoid+softmax instead of softmax normalization (ablation)A²U dynamic-cs matting network · Composition-1k testSAD: 36.40 -> 35.86Improved

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