pointpillars 开源项目推荐（32 个）

zhulf0804 /PointPillars

一个用于 3D LiDAR（KITTI）检测的简单 PointPillars PyTorch 实现。

PaddlePaddle /Paddle3D

基于PaddlePaddle的3D计算机视觉开发工具包。它支持点云物体检测、分割和单眼三维物体检测模型。

SmallMunich /nutonomy_pointpillars

Convert pointpillars Pytorch Model To ONNX for TensorRT Inference

AgentMaker /PAPC

PAPC is a deep learning for point clouds platform based on pure PaddlePaddle

anshulpaigwar /Frustum-Pointpillars

Frustum-PointPillars: A Multi-Stage Approach for 3D Object Detection using RGB Camera and LiDAR

Loahit5101 /PointPillars-Camera-LiDAR-Fusion

Implementation of PointPillars Network with LiDAR-camera fusion for 3D Object Detection in Autonomous Driving.

SCRN-VRC /Point-Pillars-in-UnityCG

Implementation of PointPillars paper in Unity for VRChat

ksm26 /ROS-based-3D-detection-Tracking

Deployment of 3D-Detection and Tracking pipeline in simulation based on rosbags and real-time.

shangjie-li /pointpillars

Implementation of PointPillars in PyTorch for KITTI 3D Object Detetcion

aydnzn /Enhancing-LiDAR-based-3D-Object-Detection

This repository contains a Master's Thesis on enhancing LiDAR-based 3D Object Detection in autonomous vehicles using synthetic data. The research explores the use of the Ansys AVxcelerate Sensors Simulator (AVX) to create synthetic point clouds. It includes a hybrid training approach, experimental design, results, and related scripts.

LioEinaudi /CUDA-Accelerated-LiDAR-PointPillars-Preprocessing

C++/CUDA PointPillars LiDAR preprocessing pipeline with KITTI loading, pillar scatter, BEV pseudo-image generation, tests, and CUDA range filtering.

surabhi96 /pointpillars

working repository

QkkooQ /4D-RADAR

4D mmWave radar + monocular camera 3D perception prototype built with PyTorch.

muhammadmahadazher /laserperception

Reproducible 3D LiDAR detection with TensorRT FP16, exact deterministic voxelization, and ROS 2.

jitrc /object-detection

Experimental dockerized workspace for 2D/3D object detection, segmentation, tracking, anything related to perception

aydnzn /Master-Thesis-Supplements

This repository contains supplementary materials for a Master Thesis project on efficient object detection. It includes a series of video demonstrations showcasing the performance of networks trained and tested on synthetic (AVX) and real-world (KITTI) data sets.

Sweetest316 /ros2_pointCloud_demo

ROS2 PointPillars inference demo using OpenPCDet and pybind11 for LiDAR point cloud object detection.

mcocheteux /PseudoCal

Initialisation-free camera–LiDAR self-calibration via pseudo-LiDAR BEV pillars (BMVC 2023). Recovers extreme decalibration — up to ±180° yaw — in one shot, then refines through a 3-stage coarse-to-fine cascade.

GunjanKholapure /second.pytorch

PointPillars for Argoverse object detection

sherifabdelrady /autoperception

3D BEV object detection · mAP 0.624 on nuScenes · LiDAR-camera fusion · PyTorch