AI Paper Atlas

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

GUIDE: Gaussian Unified Instance Detection for Enhanced Obstacle Perception in Autonomous Driving

Chunyong Hu, Qi Luo, Jianyun Xu, Song Wang, Qiang Li, Sheng Yang

AAAI 2026 · Detection and segmentation › 3D detection and driving · approach: LiDAR perception for driving · 0 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
increase Gaussians per instance from 16 to 48 (ablation)GUIDE · nuScenes valmAP_occ (10 classes): 18.94 -> 22.81Improved
increase Gaussians per instance from 16 to 48 (ablation)GUIDE · nuScenes valmIoU: 22.79 -> 25.39Improved
increase Gaussians per instance from 16 to 48 (ablation)GUIDE · nuScenes vallatency: 205ms -> 373msHurt
increase Gaussians per instance from 16 to 48 (ablation)GUIDE · nuScenes valGPU memory: 2926 MB -> 3434 MBHurt
add instance bank (temporal fusion) (ablation)GUIDE · nuScenes valmAP_occ (10 classes): 16.42 -> 22.81Improved
add instance bank (temporal fusion) (ablation)GUIDE · nuScenes valmIoU (10 fg classes): 21.51 -> 25.39Improved
increase candidate instances from 300 to 900 (ablation)GUIDE · nuScenes valmAP_occ (10 classes): 18.94 -> 22.81Improved
increase candidate instances from 300 to 900 (ablation)GUIDE · nuScenes valmIoU (10 fg classes): 22.68 -> 25.39Improved
use instance Gaussian representation instead of SparseOccGUIDE · nuScenes valmAP_occ (8 classes): 14.40 -> 21.61Improved
use instance Gaussian representation instead of SparseOccGUIDE · nuScenes valGPU memory: 5424 MB -> 3434 MBImproved
use instance Gaussians instead of SparseOcc for trackingGUIDE vs SparseDrive-S · nuScenes valIDS: 886 -> 516Improved
GUIDE vs SparseOcc on barriers and traffic conesGUIDE · nuScenes val (barrier, traffic cone)mIoU: 28.99 -> 30.96Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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