AI Paper Atlas

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

Accelerate 3D Object Detection Models via Zero-Shot Attention Key Pruning

Lizhen Xu, Xiuxiu Bai, Xiaojun Jia, Jianwu Fang, Shanmin Pang

ICCV 2025 · Detection and segmentation › 3D detection and driving · approach: Unified LiDAR 3D detection · 2 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 max instead of mean of classification vector for key importance (ablation)tgGBC on StreamPETR-VovNet · nuScenes valmAP (%): 44.12% -> 48.55%Improved
use max instead of min of classification vector for key importance (ablation)tgGBC on StreamPETR-VovNet · nuScenes valmAP (%): 34.88% -> 48.55%Improved
use max instead of mean of classification vector for key importance (ablation)tgGBC on OPEN-R101 · nuScenes valmAP (%): 49.57% -> 51.57%Improved
use max instead of min of classification vector for key importance (ablation)tgGBC on OPEN-R101 · nuScenes valmAP (%): 47.65% -> 51.57%Improved
weight attention by classification scores in key importance (ablation)tgGBC · nuScenes valmAP and NDSImproved
use n=2 tgGBC layers instead of n=1 (ablation)tgGBC on StreamPETR-VovNet · nuScenes valmAP (%): 47.84% -> 48.55%Improved
use n=2 tgGBC layers instead of n=1 (ablation)tgGBC on StreamPETR-VovNet · nuScenes valdecoder time (ms): 31.61 -> 34.98Hurt
use n=5 instead of n=3 tgGBC layers (ablation)tgGBC on StreamPETR-VovNet · nuScenes valmAP (%): 48.21% -> 48.92%Mixed
increase number of pruned keys r (ablation)tgGBC · nuScenes valmAP and NDSMixed
replace ToMe with tgGBCStreamPETR-R50-704x256 · nuScenes valmAP (%): 36.12% -> 37.93%Improved
replace ToMe with tgGBCOPEN-VovNet-800x320 · nuScenes valmAP (%): 49.21% -> 52.12%Improved
add tgGBC pruning to ToC3DToC3D · nuScenes valdecoder time (ms): 77.92 -> 39.17Improved

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