AI Paper Atlas

Research map / Point clouds / Point cloud classification

Sequential Point Clouds: A Survey

Haiyan Wang, Yingli Tian

TPAMI 2024 · Point clouds › Point cloud classification · approach: Local feature aggregation networks · 23 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 multiple frames instead of a single frameMVF++ · Waymo Open val Vehicles (IoU=0.7)3D AP: 74.64 -> 79.73Improved
use offboard multi-frame detection instead of single-frame MVF++Qi et al. · Waymo Open val Vehicles (IoU=0.7)3D AP: 74.64 -> 84.50Improved
use graph-based multi-frame detector instead of best single-frameYin et al. · nuScenes detectionmean AP: 42.0 -> 45.4Improved
use convolution-based multi-frame detector instead of single-frame PointPillarsWhat you see · nuScenes detectionmean AP: 30.5 -> 35.0Improved
use point-based network instead of voxel-based network for scene flowFlowNet3D vs VoxFlowNet · FlyingThings3D scene flowEPE (m): 0.2971 -> 0.1694Improved
train scene flow without supervisionPointPWCNet · FlyingThings3D scene flowEPE (m): 0.0588 -> 0.1246Hurt
train on FlyingThings3D and test on real KITTI without retrainingPV-RAFT · KITTI scene flowEPE (m): 0.0461 -> 0.056No clear effect
use 4 scans instead of 1 scan as inputAygun et al. · SemanticKITTI 4D panoptic valLSTQ: 43.88 -> 56.89Improved
propagate instance IDs with scene flow instead of multi-object trackingAygun et al. · SemanticKITTI 4D panoptic valLSTQ: 40.18 -> 43.88Improved
use multiple frames instead of one frame with sparse voxel network4D MinkNet14 · Synthia 4D semantic segmentationmIoU: 76.24 -> 77.46Improved
increase PSTNet temporal kernel l from 1 to 3PSTNet · Synthia 4D semantic segmentationmIoU: 80.79 -> 82.24Improved
use RNN-based instead of convolution/graph-based detectors for temporal modelingRNN-based detectors · nuScenes detectionmean APHurt

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