AI Paper Atlas

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

A Semantic Space is Worth 256 Language Descriptions: Make Stronger Segmentation Models with Descriptive Properties

Junfei Xiao, Ziqi Zhou, Wenxuan Li, Shiyi Lan, Jieru Mei, Zhiding Yu, Bingchen Zhao, Alan Yuille, Yuyin Zhou, Cihang Xie

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 3 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
replace category-level label space with property-level label spaceViT-Adapter DeiT-B · Pascal ContextmIoU: 53.3 -> 58.2Improved
replace category-level label space with property-level label spaceViT-Adapter DeiT-B · BDDmIoU: 60.7 -> 65.7Improved
replace category-level label space with property-level label spaceViT-Adapter DeiT-B · ADE20KmIoU: 48.4 -> 49.0Improved
add ProLab property-level label space to DeepLabv3+ and SegFormerDeepLabv3+ ResNet50 / SegFormer MiT-B1 · ADE20K, Cityscapes, BDDmIoU: DeepLabv3+ ADE20K 42.7 -> 43.6; SegFormer ADE20K 41.5 -> 42.3Improved
cluster description embeddings into properties (vs no clustering) (ablation)ProLab · ADE20KmIoU: 30.2 -> 48.3Improved
vary number of clusters (64, 128, 256, 512) (ablation)ProLab · ADE20KmIoU: 47.8 / 48.0 / 48.3 / 47.6No clear effect
use property-level embedding space instead of category-level embedding space (BGE-B) (ablation)ProLab · ADE20KmIoU: 47.7 -> 49.0Improved
use BERT or T5 text encoders instead of sentence embedding model (ablation)ProLab · ADE20KmIoU: 47.7 (BGE-B) -> 42.0 (T5-B), 40.8 (BERT-B)Hurt
use the crafted Aligned description prompt instead of Naive prompt (ablation)ProLab with GPT-3.5 · ADE20KmIoU: 47.5 -> 47.7Improved
use GPT-3.5 instead of LLAMA2-7B for descriptions (ablation)ProLab · ADE20KmIoU: 47.0 -> 47.5Improved
apply sigmoid with cosine similarity loss instead of BCE (ablation)ProLab · ADE20KmIoU: 47.4 -> 47.7Improved
cosine similarity loss without sigmoid vs BCE (ablation)ProLab · ADE20KmIoU: 47.4 -> 47.3No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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