AI Paper Atlas

Research map / Detection and segmentation / CNN features

CC-SAM: Enhancing SAM with Cross-feature Attention and Context for Ultrasound Image Segmentation

Shreyank N Gowda, David A Clifton

ECCV 2024 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation

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 frozen pretrained ResNet50 branch (ablation)SAMUS baseline · TN3KDice (%): 84.45 -> 84.91Improved
add variational attention fusion (VAF) (ablation)SAMUS baseline · TN3KDice (%): 84.45 -> 85.55Improved
add GPT-4 text prompts instead of point prompts (ablation)SAMUS baseline · BUSIDice (%): 85.77 -> 87.92Improved
add text prompts on top of ResNet50 + VAF (ablation)CC-SAM · TN3KDice (%): 85.72 -> 87.11Improved
add VAF and text prompt on top of ResNet50 (ablation)CC-SAM · BUSIDice (%): 88.45 -> 89.51Improved
CC-SAM versus SAMUS on foundation-model benchmarkCC-SAM · TN3K (US30K)Dice (%): 83.05 -> 85.59Improved
CC-SAM versus FAT-Net on CAMUS-MYO Hausdorff distanceCC-SAM · CAMUS-MYOHausdorff distance: 15.93 vs 16.11Hurt
CC-SAM on unseen datasets (generalization)CC-SAM · unseen DDTI, UDIAT, HMC-QUseen-to-unseen performance dropImproved

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