AI Paper Atlas

Research map / Detection and segmentation / CNN features

PANDORA: A Panoramic Detection Dataset for Object with Orientation

Hang Xu, Qiang Zhao, Yike Ma, Xiaodong Li, Peng Yuan, Bailan Feng, Chenggang Yan, Feng Dai

ECCV 2022 · Detection and segmentation › CNN features · approach: Box regression and loss design · 24 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
use rotated bounding FoV (RBFoV) instead of BFoV as the box representation (ablation)Sph-CenterNet · PANDORAAP50: 20.3 -> 21.4 (+1.1)Improved
use RBFoV instead of BFoV to reduce overlap of neighbouring boxes (ablation)Sph-CenterNet · PANDORA overlap analysisA: affected instances ratio: 0.23 -> 0.11Improved
use RBFoV instead of BFoV for overlap miss share (ablation)Sph-CenterNet · PANDORA overlap analysisC: overlap misses / overlap cases: 0.27 -> 0.18Improved
add all proposed components (new heatmap, rotation augmentation, angle regression) to Sph-CenterNet (ablation)Sph-CenterNet · PANDORA ablationAP50: 19.9 -> 22.7 (+2.8)Improved
add a partial subset of components to Sph-CenterNet (first combination in Table 4) (ablation)Sph-CenterNet · PANDORA ablationAP50: 19.9 -> 20.5 (+0.6)Improved
add a partial subset of components to Sph-CenterNet (second combination in Table 4) (ablation)Sph-CenterNet · PANDORA ablationAP50: 19.9 -> 21.6 (+1.7)Improved
add a partial subset of components to Sph-CenterNet (third combination in Table 4) (ablation)Sph-CenterNet · PANDORA ablationAP50: 19.9 -> 21.4 (+1.5)Improved
R-CenterNet versus Sph-CenterNet on RBFoV ground truthR-CenterNet · PANDORA test (RBFoV GT)AP: 5.5 -> 7.3Improved
two-stage detectors versus one-stage Sphere-SSDReprojection R-CNN · PANDORA test (RBFoV GT)AP50: 12 -> 16.6Improved

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