AI Paper Atlas

Research map / Point clouds / Segmentation and detection

Just Add $100 More: Augmenting Pseudo-LiDAR Point Cloud for Resolving Class-imbalance Problem

Mincheol Chang, Siyeong Lee, Jinkyu Kim, Namil Kim

NeurIPS 2024 · Point clouds › Segmentation and detection · approach: LiDAR voxel and pillar detectors · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes PGT-Aug, which turns low-cost surround-view videos of miniatures and public videos into NeRF/Gaussian-splat-based pseudo-LiDAR objects for minority-class augmentation in LiDAR 3D detection, with domain alignment and map-aware placement.

Problem

Class imbalance (long-tailed distribution) in LiDAR-based 3D object detection datasets, where minority classes have few samples and existing copy-paste augmentation has limited sample diversity and restricted placement.

Approach

Reconstruct volumetric 3D instances of minority-class objects from multi-view video (miniatures bought for about $100 or public videos) with Plenoxels/Gaussian Splatting, convert them to LiDAR-like pseudo point clouds via range-view rearrangement and a CycleGAN-based intensity estimator, and place them using ground-plus-map-aware insertion areas.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add PGT-Aug pseudo-LiDAR samples with ground+map placement instead of Real-AugCP-Voxel · nuScenes valmAP: 62.27 -> 63.52Improved
Add PGT-Aug pseudo-LiDAR samples with ground+map placement instead of Real-AugCP-Voxel · nuScenes valNDS: 68.39 -> 69.11Improved
Replace Real-Aug with PGT-AugCP-Voxel · nuScenes testmAP: 65.8 -> 67.1Improved
Replace Real-Aug with PGT-AugCP-Voxel · nuScenes testNDS: 71.3 -> 72.3Improved
Replace GT-Aug with PGT-AugCP-Voxel · Lyft valmAP: 13.84 -> 14.01Improved
Replace GT-Aug with PGT-AugSECOND · KITTI valmAP: 68.5 -> 70.1Improved
Use azimuth resolution 3600 px instead of 1080 px (ablation)Plenoxel pseudo-LiDAR generation · nuScenes val pseudo-LiDAR qualityAvg. FID Score: 7.7 -> 14.2Hurt
Add group intensity loss (ablation)Plenoxel pseudo-LiDAR generation without RGB features · nuScenes val pseudo-LiDAR qualityAvg. FID Score: 8.2 -> 8.3Hurt
Add RGB features (ablation)Plenoxel pseudo-LiDAR generation without group intensity loss · nuScenes val pseudo-LiDAR qualityAvg. FID Score: 8.2 -> 8.3Hurt
Replace Plenoxel with Gaussian Splatting (ablation)PGT-Aug · nuScenes val pseudo-LiDAR qualityAvg. FID Score: 7.7 -> 7.1Improved
Add ray tracing (ablation)PGT-Aug with CP-Voxel · nuScenes valmAP: 63.5 -> 63.8Improved
Add ray tracing (ablation)PGT-Aug with Transfusion-L · nuScenes valmAP: 64.2 -> 65.0Improved

Surprising findings

Limitations

Experimental setup

Datasets: nuScenes KITTI Lyft A2D2

Benchmarks: nuScenes KITTI Lyft

Metrics: AP mAP NDS FID

Baselines: GT-Aug Real-Aug CP-Voxel CP-Pillar SECOND Transfusion-L VoxelNeXt

Builds on: GT-Aug Real-Aug Plenoxels 3D Gaussian Splatting CycleGAN

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