AI Paper Atlas

Research map / Detection and segmentation / CNN features

Structural and Statistical Texture Knowledge Distillation and Learning for Segmentation

Deyi Ji, Feng Zhao, Hongtao Lu, Feng He Wu, Jieping Ye

TPAMI 2025 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 11 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 adversarial learning to output-probability response KD (ablation)response KD · Cityscapes valmIoU: 70.51 -> 72.47Improved
add structural texture KD to response KD (ablation)SSTKD student R18 · Cityscapes valmIoU: 72.47 -> 74.80Improved
add statistical texture KD to response KD (ablation)SSTKD student R18 · Cityscapes valmIoU: 72.47 -> 75.89Improved
add both textures (ablation)SSTKD student R18 · Cityscapes valmIoU: 72.47 -> 76.65Improved
contourlet instead of FFT for structural texture (ablation)structural KD · Cityscapes valmIoU: 72.78 -> 74.80Improved
contourlet instead of wavelet (ablation)structural KD · Cityscapes valmIoU: 73.13 -> 74.80Improved
TIEM instead of GLCM for statistical texture (ablation)statistical KD · Cityscapes valmIoU: 72.97 -> 75.89Improved
add denoising step (ablation)statistical texture KD · Cityscapes valmIoU: 75.06 -> 75.89Improved
add adaptive sampling (ablation)statistical texture KD · Cityscapes valmIoU: 75.37 -> 75.89Improved
mild uncertainty-biased sampling (k=2, beta=0.7) vs k=1 beta=0 (ablation)adaptive sampler · Cityscapes valmIoU: 73.70 -> 75.89Improved
heavier biased sampling (k=8, beta=1.0) vs k=2 beta=0.7 (ablation)adaptive sampler · Cityscapes valmIoU: 75.89 -> 75.26Hurt
add SSTKD over CWD distillationPSPNet-R18* · Cityscapes valmIoU: 75.90 -> 76.65Improved

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