AI Paper Atlas

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

Bamboo: Building Mega-Scale Vision Dataset Continually with Human–Machine Synergy

Yuanhan Zhang, Qinghong Sun, Yichun Zhou, Zexin He, Zhenfei Yin, Kun Wang, Lu Sheng, Yu Qiao, Jing Shao, Ziwei Liu

IJCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Remote sensing segmentation and change · 22 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
pretrain on Bamboo-CLS instead of CLIP WIT 400MViT-B/16 · downstream linear probe averageaccuracy: 85.6 -> 91.8Improved
pretrain on Bamboo-CLS instead of CLIP WIT 400MRN50 · downstream linear probe averageaccuracy: 79.1 -> 86.0Improved
pretrain on Bamboo-DET instead of Objects365RN50-FPN · PASCAL VOCAP50: 86.4 -> 87.5Improved
pretrain on Bamboo-DET instead of Objects365RN50-FPN · CityPersonsMR: 14.7 -> 12.6Improved
pretrain on Bamboo-DET instead of Objects365RN50-FPN · COCOmmAP: 39.3 -> 43.9Improved
use Bamboo-CLS pretraining instead of CLIPRN50 · zero-shot classification averageaccuracy: 70.0 -> 76.0Improved
fine-tune Bamboo-CLS instead of IG-1B SWSLRN50 · fine-tuning averageaccuracy: 86.8 -> 87.3Improved
fine-tune Bamboo-CLS instead of ImageNet22K BiT-MRN50 · fine-tuning averageaccuracy: 86.0 -> 87.3Improved
pretrain on Bamboo instead of JFT-300MRN50 · ObjectNetaccuracy: 37.6 -> 38.8Improved
pretrain on Bamboo instead of ANN-1.3BViT-B/16 · ObjectNetaccuracy: 50.7 -> 53.9Improved
label sparse images instead of crowded images under same proposal budget (ablation)Objects365 subsets · PASCAL VOCAP50: 85.1 -> 86.3Improved
sample randomly instead of keeping dense crowded images (same budget) (ablation)Objects365 subsets · PASCAL VOCAP50: 85.1 -> 85.8Improved

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