AI Paper Atlas

Research map / Detection and segmentation / CNN features

Few-Shot End-to-End Object Detection via Constantly Concentrated Encoding across Heads

Jiawei Ma, Guangxing Han, Shiyuan Huang, Yuncong Yang, Shih-Fu Chang

ECCV 2022 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 19 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 the full CoCo regularization (class encodings, SupCT on heads 2-6, background negatives) (ablation)Sparse-RCNN finetuning · MS COCO 30-shotnAP: 16.3 -> 20.6Improved
add class encodings to learnable proposal vectors (ablation)Sparse-RCNN finetuning · MS COCO 30-shotnAP: 16.3 -> 17.3Improved
add SupCT among object encodings of heads 2 to 6 (ablation)Sparse-RCNN finetuning · MS COCO 30-shotnAP: 16.3 -> 19.2Improved
combine class encodings with SupCT on heads 2-6 (ablation)CoCo-RCNN · MS COCO 30-shotnAP: 19.2 -> 20.1Improved
add background object encodings as negatives in SupCT (ablation)CoCo-RCNN · MS COCO 30-shotnAP: 20.1 -> 20.6Improved
use soft sampling from neighbouring heads instead of hard label assignment across heads (ablation)CoCo regularization · VOC 10-shotnAP: 13.0 -> 47.2; 18.5 -> 47.2Improved
use supervised contrastive loss instead of soft-label distillation across heads (ablation)CoCo regularization · VOC 10-shotnAP: 43.1 -> 47.2Improved
use SupCT over positives from several heads instead of plain contrastive (ablation)CoCo regularization · VOC 10-shotnAP: 44.5 -> 47.2Improved
replace SupCT with IoU-weighted CPE loss (ablation)CoCo regularization · VOC 10-shotnAP: 46.0 -> 47.2No clear effect
sample positives using two neighbouring heads instead of previous-head labels only (ablation)CoCo regularization · VOC 10-shotnAP: 43.3 -> 47.2Improved
use separate projectors per head instead of a shared projector (ablation)CoCo-RCNN · VOC 10-shot / MS COCO 30-shotnAP50 / nAP: 64.0 -> 65.5; 19.8 -> 20.6Improved
add CoCo regularization to large-scale training from scratchSparse-RCNN · COCO large-scaleAP: 46.3 -> 47.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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