AI Paper Atlas

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

PSALM: Pixelwise Segmentation with Large Multi-modal Model

Zheng Zhang, yeyao ma, Enming Zhang, Xiang Bai

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 54 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
feed mask tokens through the LLM instead of directly to mask generator (ablation)PSALM · RefCOCO valcIoU: 74.3 -> 76.1Improved
feed mask tokens through the LLM instead of directly to mask generator (ablation)PSALM · A150-OVmAP: 8.2 -> 9.3Improved
place mask tokens as suffix rather than prefix (ablation)PSALM · A150-OVmAP: 7.8 -> 9.3Improved
decouple mask prediction and classification (ablation)PSALM · COCO Semantic SegmentationmIoU: 42.7 -> 66.5Improved
use 100 mask proposals instead of a single mask (ablation)PSALM · RefCOCO valcIoU: 75.3 -> 76.5Improved
use [REF] token for sentence condition instead of avg_pooling (ablation)PSALM · RefCOCO valcIoU: 75.3 -> 76.1Improved
use avg_pooling for category condition instead of [REF] (ablation)PSALM · A150-OVmAP: 8.4 -> 9.3Improved
add vision-language alignment stage (ablation)PSALM · RefCOCO valcIoU: 71.7 -> 76.1Improved
add vision-language alignment stage (ablation)PSALM · COCO-PanPQ: 54.9 -> 55.1No clear effect
joint training across tasks instead of task-specific training (ablation)PSALM · RefCOCO valcIoU: 76.5 -> 83.6Improved
joint training across tasks instead of task-specific training (ablation)PSALM · COCO-PanPQ: 55.6 -> 55.9No clear effect
joint training across tasks instead of task-specific training (ablation)PSALM · A150-OVmAP: 7.0 -> 9.0Improved

Later papers that compared with this method

7 later accepted papers report PSALM as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.

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