AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Improving Robustness of Semantic Segmentation to Motion-Blur Using Class-Centric Augmentation

Aakanksha, A. N. Rajagopalan

CVPR 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 14 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
Apply class-centric motion-blur augmentation instead of the strongest baselineDeepLabv3+ ResNet-50 · PASCAL VOC L1, L2, L3mIoU: 2.3%, 3.5% and 3.2%Improved
Apply class-centric motion-blur augmentation instead of the strongest baselineDeepLabv3+ ResNet-50 · Cityscapes L1, L2, L3mIoU: 1.9%, 3.4% and 3.1%Improved
Apply CCMBA rather than the best baseline at each blur levelSegformer MiT-B2 · Cityscapes L1, L2, L3mIoUNo clear effect
Apply CCMBA rather than PbN [14]DeepLabv3+ · Cityscapes-C S1, S2, S3mIoU: 2.3%, 5.2% and 9%Improved
Apply CCMBA rather than pre-trained SegformerSegformer · Cityscapes-C S2, S3mIoU: 1.7%, 5.1 %Improved
Apply CCMBA rather than MBASegformer · Cityscapes-C clean and S3mIoU: 2%, 0.5%Improved
Initialize CCMBA training from clean pre-trained weights rather than random initialization (ablation)DeepLabv3+ · PASCAL VOC clean, L1, L2, L3mIoU: Random: 75.5, 73.6, 69.6, 61.5; Pre-trained: 76.5, 74.6, 72.1, 66.0Improved
Use a larger backbone (ablation)CCMBA with DeepLabv3+ · PASCAL VOC clean and L1, L2, L3mIoU: MobileNetV2: 69.3, 68.2, 65.6, 61.5; ResNet-50: 76.3, 75.3, 72.5, 68.6; ResNet-101: 77.6, 76.3, 75.2, 71.3Improved
Train on all three blur levels rather than one blur level (ablation)CCMBA with DeepLabv3+ MobileNetV2 · PASCAL VOC clean and L1, L2, L3mIoU: Clean: 69.3; L1: 68.2; L2: 66.6; L3: 61.5Improved
Use only clean training images rather than blurred augmentation (ablation)DeepLabv3+ MobileNetV2 · PASCAL VOC L1, L2, L3mIoU: 60.3, 44.4, 29.0Hurt
Train using only L1 blur rather than all three blur levels (ablation)CCMBA with DeepLabv3+ MobileNetV2 · PASCAL VOC L1, L2, L3mIoU: 65.1, 58.7, 39.1Hurt
Train using only L2 blur rather than all three blur levels (ablation)CCMBA with DeepLabv3+ MobileNetV2 · PASCAL VOC L1, L2, L3mIoU: 65.6, 63.2, 53.5Hurt

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