AI Paper Atlas

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

Going Denser with Open-Vocabulary Part Segmentation

Peize Sun, Shoufa Chen, Chenchen Zhu, Fanyi Xiao, Ping Luo, Saining Xie, Zhicheng Yan

ICCV 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 68 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 parsed part annotations of novel objects (Parsed IN-S11) to training (ablation)VLPart trained on Pascal Part · PartImageNet val, 40 parts (cross-dataset)mAP: 4.5 -> 7.8Improved
add image-level IN-S11 labels without parsing (ablation)VLPart trained on Pascal Part · PartImageNet val, 40 partsmAP: 4.5 -> 5.4Improved
add LVIS and PACO part-level data (ablation)VLPart trained on Pascal Part · PartImageNet val, 40 partsmAP: 4.5 -> 7.8Improved
add VOC object-level data (ablation)Pascal Part base-part model · Pascal Part novel partsAP50: 3.7 -> 5.9Improved
add image-level IN-S20 labels on top of object data (ablation)VLPart on Pascal Part base split · Pascal Part novel partsAP50: 5.9 -> 3.1 (novel); 36.8 -> 37.5 (all)No clear effect
add parsed IN-S20 part annotations (ablation)VLPart on Pascal Part base split · Pascal Part novel partsAP50: 3.1 -> 11.0Improved
joint training on LVIS, PACO, Pascal Part, PartImageNet (no extra image data) (ablation)VLPart ResNet50 · PACO and PartImageNet vs dataset-specificAP: PACO 9.3 vs 10.6; Pascal Part 22.6 vs 19.4Mixed
add IN and Parsed IN to joint training (ablation)VLPart ResNet50 joint · PartImageNet, Pascal Part, PACOAP: PartImageNet 29.1 -> 31.6; Pascal Part 22.6 -> 24.0; PACO 9.3 -> 9.6Improved
scale backbone from ResNet50 to Swin-B Cascade Mask R-CNNjoint VLPart · PACOAP: 9.6 -> 15.6Improved
use max-size alignment of image labels instead of parsing (ablation)VLPart cross-dataset · PartImageNet, 40 partsmAP: 5.4 -> 5.3Hurt
use max-score alignment instead of parsing (ablation)VLPart cross-dataset · PartImageNet, 40 partsmAP: 5.4 -> 6.0 (parsing: 7.8)Improved
swap prompt template to '[part] of a [object]' (ablation)fully supervised part detector · Pascal Part vs PartImageNetAP: Pascal 19.1 -> 18.4; PartImageNet 29.7 -> 29.9Mixed

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