AI Paper Atlas

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

ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation

Mengcheng Lan, Chaofeng Chen, Yiping Ke, Xinjiang Wang, Litong Feng, Wayne Zhang

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 70 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 adaptive normalization and masking to proxy attention (ablation)ProxyCLIP (CLIP-ViT-B/16, DINO B/8) · 8 benchmarks avgmIoU: 25.0 -> 42.3Improved
add masking on top of normalization (ablation)ProxyCLIP · 8 benchmarks avgmIoU: 29.5 -> 42.3Improved
use VFM feature similarity as proxy attention instead of VFM's own attention scores (ablation)ProxyCLIP · 8 benchmarks avgmIoU: 24.1 -> 42.3 (DINO B/8)Improved
use DINO with patch size 8 instead of 16 (ablation)ProxyCLIP · 8 benchmarks avg (CLIP-ViT-B/16)mIoU: 41.1 -> 42.3Improved
use larger OpenCLIP-ViT-H/14 backbone instead of CLIP-ViT-B/16 (ablation)ProxyCLIP · 8 benchmarks avgmIoU: 42.3 -> 44.4Improved
scale CLIP backbone from B/16 to L/14 or H/14 in training-free SCLIPSCLIP · 8 benchmarks avgmIoU: 38.2 -> 29.0 (L/14), 29.1 (H/14)Hurt
swap VFM among DINO, DINOv2, SAM (ablation)ProxyCLIP · 8 benchmarks avg (CLIP-ViT-B/16)mIoU: 41.1 / 41.1 / 40.8No clear effect
use Stable Diffusion UNet features as VFM instead of DINO B/16 (ablation)ProxyCLIP · 8 benchmarks avg (CLIP-ViT-B/16)mIoU: 41.1 -> 41.8Improved
use fixed hard-mask threshold instead of adaptive normalization (ablation)ProxyCLIP · COCO StuffmIoU: DINO best 25.8 at 0.4 vs 26.5 adaptiveMixed
use VFM embeddings instead of CLIP q-q attention in PAM (ablation)ProxyCLIP · 8 benchmarks avg (CLIP-ViT-B/16)mIoU: 38.2 -> 42.3Improved
add DINO proxy attention to CLIPCLIP-ViT-B/16 · 8 benchmarks avgmIoU: 11.7 -> 42.3Improved

Later papers that compared with this method

19 later accepted papers report ProxyCLIP as a baseline in their results tables; 19 of them report a better number on most of the results they share with it.

11 more not listed.

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