AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

FocalFormer3D: Focusing on Hard Instance for 3D Object Detection

Yilun Chen, Zhiding Yu, Yukang Chen, Shiyi Lan, Anima Anandkumar, Jiaya Jia, Jose M. Alvarez

ICCV 2023 · Detection and segmentation › 3D detection and driving · approach: Depth-based and label-free 3D detection · 129 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 one extra hard-instance-probing stage at 200 queries (ablation)FocalFormer3D · nuScenes valmAP: 65.3 -> 66.0Improved
increase single-stage queries from 200 to 600 without HIP (ablation)FocalFormer3D · nuScenes valmAP: 65.3 -> 65.4No clear effect
use 3 HIP stages with 600 queries instead of 1 stage (ablation)FocalFormer3D · nuScenes valmAP: 65.4 -> 66.5Improved
add pooling-based positive masking (ablation)FocalFormer3D · nuScenes valmAP: 65.3 -> 66.5Improved
use point masking instead of no masking (ablation)FocalFormer3D · nuScenes valmAP: 65.3 -> 65.9Improved
use box masking instead of pooling-based masking (ablation)FocalFormer3D · nuScenes valmAP: 66.5 -> 66.1Hurt
add multi-stage heatmap encoder without second-stage refinement (ablation)center-based detector · nuScenes valmAP: 63.1 -> 63.3No clear effect
add box-pooling RoI refinement on top of multi-stage heatmaps (ablation)FocalFormer3D · nuScenes valmAP: 63.3 -> 65.1Improved
use full box-level deformable decoder (box pooling plus cross attention) (ablation)FocalFormer3D · nuScenes valmAP: 65.1 -> 66.5Improved
add rescoring only without box regression (ablation)FocalFormer3D · nuScenes valmAP: 63.1 -> 66.1Improved
use box-level query embedding instead of point-level (ablation)FocalFormer3D decoder · nuScenes valmAP: 65.9 -> 66.5Improved
use deformable attention instead of full cross attention (ablation)FocalFormer3D decoder · nuScenes val, 600 querieslatency: 26.2ms (full, 2 layers) vs 17.0ms (deformable, 6 layers)Improved

Later papers that compared with this method

4 later accepted papers report FocalFormer3D as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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