AI Paper Atlas

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

MARIS: Marine Open-Vocabulary Instance Segmentation

Bingyu Li, Feiyu Wang, Da Zhang, Zhiyuan Zhao, Junyu Gao, Xuelong Li

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary 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 GPEM geometric prior module to the baseline (ablation)ConvNeXt-L OV segmenter · MARIS in-domain, overall classesmAP: 52.17 -> 54.99Improved
add SAIM semantic alignment module to the baseline (ablation)ConvNeXt-L OV segmenter · MARIS in-domain, overall classesmAP: 52.17 -> 55.27Improved
combine GPEM and SAIM (ablation)ConvNeXt-L OV segmenter · MARIS in-domain, OV classesmAP: 50.99 -> 54.02Improved
add underwater prompts instead of generic templates (ablation)SAIM, ConvNeXt-B · MARIS in-domain, overall classesmAP: 42.91 -> 43.89Improved
add similarity-based template selection on top of underwater prompts (ablation)SAIM, ConvNeXt-B · MARIS in-domain, overall classesmAP: 43.89 -> 44.54Improved
use adaptive alpha fusion instead of MLP fusion (ablation)GPEM, ConvNeXt-B · MARIS in-domain, overall classesmAP: 43.87 -> 44.54Improved
use plain addition instead of adaptive fusion (ablation)GPEM, ConvNeXt-B · MARIS in-domain, overall classesmAP: 44.54 -> 43.52Hurt
enlarge geometry encoder from ViT-B to ViT-L (ablation)MARIS with ConvNeXt-L · MARIS in-domain vs cross-domain, overall classesmAP: in-domain 55.22 -> 56.71; cross-domain 46.18 -> 43.70Mixed
enlarge geometry encoder from ViT-B to ViT-L (ablation)MARIS with ConvNeXt-B · MARIS in-domain, overall classesmAP: 44.54 -> 44.37No clear effect
shrink geometry encoder from ViT-B to ViT-S (ablation)MARIS with ConvNeXt-B · MARIS in-domain, overall classesmAP: 44.54 -> 42.36Hurt
train on COCO and test underwater instead of in-domain trainingMARIS ConvNeXt-B · MARIS overall classesmAP: 44.37 -> 32.62Hurt
use MARIS instead of MAFT+ in cross-domainConvNeXt-L · MARIS cross-domain, overall classesmAP: 40.27 -> 46.18Improved

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