AI Paper Atlas

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

Tell Me Where I Am: Object-Level Scene Context Prediction

Xiaotian Qiao, Quanlong Zheng, Ying Cao, Rynson W.H. Lau

CVPR 2019 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 12 citations (OpenAlex)

Check 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
Use the scene-context prediction model instead of pix2pixScene context prediction · COCO-Stuff testNLLall: 2.15 -> 1.69Improved
Use the scene-context prediction model instead of pix2pixScene context prediction · COCO-Stuff testNLLobject: 2.11 -> 1.57Improved
Use the proposed model instead of pix2pixGenerated scene layouts · AMT scene-layout plausibilityplausibility score: 0.34 ± 0.12 -> 0.73 ± 0.11Improved
Use the proposed layout-discriminator representation and fine-tune itScene recognition model · SUN outdoor scene recognitionaccuracy: 49.8% -> 52.4%Improved
Add category-classifier pre-training for object embeddings (ablation)Scene context prediction · COCO-Stuff testNLLall: 1.77 -> 1.69Improved
Add category-classifier pre-training for object embeddings (ablation)Scene context prediction · COCO-Stuff testNLLobject: 1.70 -> 1.57Improved
Add shape and layout discriminators (ablation)Scene context prediction · COCO-Stuff testNLLall: 1.91 -> 1.69Improved
Add shape and layout discriminators (ablation)Scene context prediction · COCO-Stuff testNLLobject: 1.85 -> 1.57Improved
Add shape discriminator (ablation)Scene context prediction · COCO-Stuff testNLLall: 1.72 -> 1.69Improved
Add shape discriminator (ablation)Scene context prediction · COCO-Stuff testNLLobject: 1.61 -> 1.57Improved
Add layout discriminator (ablation)Scene context prediction · COCO-Stuff testNLLall: 1.88 -> 1.69Improved
Add layout discriminator (ablation)Scene context prediction · COCO-Stuff testNLLobject: 1.84 -> 1.57Improved

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