Code for a series of work in LiDAR perception, including SST (CVPR 22), FSD (NeurIPS 22), FSD++ (TPAMI 23), FSDv2, and CTRL (ICCV 23, oral).

Support PointRend, Fast_SCNN, HRNet, Deeplabv3_plus(xception, resnet, mobilenet), ContextNet, FPENet, DABNet, EdaNet, ENet, Espnetv2, RefineNet, UNet, DANet, HRNet, DFANet, HardNet, LedNet, OCNet, …

(LMNet) Moving Object Segmentation in 3D LiDAR Data: A Learning-based Approach Exploiting Sequential Data (RAL/IROS 2021)

A project demonstrating how to use CUDA-PointPillars to deal with cloud points data from lidar.

Represent Visual Objects by Point Sets

Advanced lane detection using computer vision

BEVFormer inference on TensorRT, including INT8 Quantization and Custom TensorRT Plugins (float/half/half2/int8).

Official PyTorch implementation of DD3D: Is Pseudo-Lidar needed for Monocular 3D Object detection? (ICCV 2021), Dennis Park*, Rares Ambrus*, Vitor Guizilini, Jie Li, and Adrien Gaidon.

TensorRT Plugin Autogen Tool

[ICCV 2023] Robo3D: Towards Robust and Reliable 3D Perception against Corruptions

A tutorial for getting started with the Deep Learning Accelerator (DLA) on NVIDIA Jetson

EA-LSS: Edge-aware Lift-splat-shot Framework for 3D BEV Object Detection

TensorRT deploy and PTQ/QAT tools development for FastBEV, total time only need 6.9ms!!!

天池雪浪布匹瑕疵检测，初赛线上949，排名9/2403

ImgEnhance For Obejct Detection tool

This is a pytorch type of block，including Non-local block，Simple Non-local block，GC block and all GC block; refer to paper《GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond》

This is trident-block of pytorch type,which is designed follow the share weight.

A project demonstrating Lidar related AI solutions, including three GPU accelerated Lidar/camera DL networks (PointPillars, CenterPoint, BEVFusion) and the related libs (cuPCL, 3D SparseConvolution…

Master's thesis research on 3D object detection using LiDAR and Camera data for infrastructure and railway domains, emphasizing inference optimization and utilization of temporal information for di…

Detect legs from lidar data

TensorRT Hackathon 2022 Final Competition

Pytorch framwork,read your own dataset in the form of txt or csv type

This is pytorch type of paper《Res2Net: A New Multi-scale Backbone Architecture》，including Res2net-50, Res2Next-29。

In caffe framwork，this .py can merge bn and scale layer into conv layer，which can speed up test。

This is a pytorch type of counting FLOPs and Params of nets。

2018/2019/校招/春招/秋招/算法/机器学习(Machine Learning)/深度学习(Deep Learning)/自然语言处理(NLP)/C/C++/Python/面试笔记

Refine json annotations according to COCO Style

Object detection with multi-level representations generated from deep high-resolution representation learning (HRNetV2h).

95.16% on CIFAR10 with PyTorch