AI Paper Atlas

Research map / Point clouds / Point cloud classification

Autoregressive Uncertainty Modeling for 3D Bounding Box Prediction

YuXuan Liu, Nikhil Mishra, Maximilian Sieb, Yide Shentu, Pieter Abbeel, Xi Chen

ECCV 2022 · Point clouds › Point cloud classification · approach: Local feature aggregation networks · 9 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
predict box parameters autoregressively instead of independent discrete bins (beam search) (ablation)Discrete box head · COB-3DIoU: 0.5232 -> 0.6296Improved
predict box parameters autoregressively instead of independent discrete bins (beam search) (ablation)Discrete box head · COB-3Derr_quat: 0.2926 -> 0.1598Improved
use autoregressive beam search instead of Gaussian outputBox head · COB-3DIoU: 0.3169 -> 0.6296Improved
use lower quantile (0.1) instead of beam searchAutoregressive FCAF3D · SUN-RGBDIoG AP_all: 47.17 -> 60.08Improved
use lower quantile (0.1) instead of beam searchAutoregressive FCAF3D · SUN-RGBDIoU AP_all: 38.15 -> 31.06Hurt
use lower quantile (0.1) instead of beam searchAutoregressive PV-RCNN · KITTI hard (Car)IoU AP: 82.37 -> 59.75Hurt
use lower quantile (0.1) instead of beam searchAutoregressive PV-RCNN · KITTI hard (Car)IoG AP: 91.84 -> 96.02Improved
use quantile 0.1 instead of beam searchBox estimator · COB-3DIoG: 0.7877 -> 0.9723Improved
use quantile 0.1 instead of beam searchBox estimator · COB-3DIoU: 0.6296 -> 0.3821Hurt
use quantile 0.5 instead of beam searchBox estimator · COB-3DIoG: 0.7877 -> 0.8295Improved
add autoregressive head with beam search to FCAF3DFCAF3D · SUN-RGBDIoU AP_all: 37.42 -> 38.15Improved
add autoregressive head with beam search to FCAF3DFCAF3D · ScannetIoU AP_all: 43.32 -> 43.85Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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