NVIDIA-AI-IOT/Lidar_AI_Solution issues and pull requests

#284 - ImportError: Lidar_AI_Solution/CUDA-BEVFusion/bevfusion/mmdet3d/ops/bev_pool/bev_pool_ext.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZN3c104impl8GPUTrace13gpuTraceStateE

Issue - State: open - Opened by anteaglew2hcs 4 days ago

#283 - CUDA-BEVFusion: difference in the result of onnx and pytorch model

Issue - State: open - Opened by Jayden9912 9 days ago - 5 comments

#282 - spconv onnx

Issue - State: open - Opened by GEN418 12 days ago

#281 - libspconv source code

Issue - State: open - Opened by jongsik-moon 16 days ago

#280 - which ORIN chip for fps numbers?

Issue - State: closed - Opened by kurquhar 18 days ago - 1 comment

#279 - 预测我自己训练的模型

Issue - State: closed - Opened by evil-master 20 days ago - 1 comment

#278 - export-transfuser erro

Issue - State: open - Opened by evil-master 22 days ago - 1 comment

#277 - Sensor selection

Issue - State: open - Opened by suhansu about 1 month ago - 1 comment

#276 - When running the nuscenes validation set on Jeston Orin 32g, it keeps killing itself. What could be the reason?

Issue - State: closed - Opened by Xiao-Hu-Z about 1 month ago - 1 comment

#275 - How to export the SwinT model from BEVFusion?

Issue - State: open - Opened by slayerlpj about 2 months ago - 2 comments

#274 - How to run the nuscenes dataset with CUDA-BEVFusion?

Issue - State: open - Opened by onlywl9598 about 2 months ago - 1 comment

#273 - How to deploy BEVSegmentationHead

Issue - State: open - Opened by whbdmu about 2 months ago

#272 - export ONNX error

Issue - State: open - Opened by slayerlpj about 2 months ago - 1 comment

#271 - Is there some bug in TensorRT?

Issue - State: open - Opened by bujianyiwang about 2 months ago - 1 comment

#270 - Is there a version of Voxelization for float type point cloud

Issue - State: open - Opened by nuaawh about 2 months ago

#269 - export spconv onnx error

Issue - State: open - Opened by aaa1233211 about 2 months ago - 1 comment

#268 - Centerpoint Int8 quantization and deployment

Issue - State: open - Opened by zxxzxy about 2 months ago

#267 - an illegal memory access was encountered

Issue - State: open - Opened by erzhu222 2 months ago - 3 comments

#266 - The outputs is all NAN

Issue - State: open - Opened by chenzb0712 2 months ago

#265 - Support for H100 and sm_90

Issue - State: open - Opened by maueki 2 months ago - 2 comments

#264 - Could libspconv provide the IOBinding feature like TensorRT?

Issue - State: closed - Opened by ThomasMouse 2 months ago - 1 comment
Labels: Feature Request

#263 - The number of voxels in Centerpoint and CUDA-CenterPoint did not match.

Issue - State: open - Opened by mx2013713828 2 months ago

#262 - Does the point cloud range need to be symmetrical from zero?

Issue - State: open - Opened by san9569 2 months ago - 1 comment

#261 - Failed to build model model/resnet50int8/build/head.bbox.layernormplugin.plan.

Issue - State: open - Opened by Happynewyear000 3 months ago - 9 comments

#260 - Only lidar or camera

Issue - State: open - Opened by MarioUncle-Wang 3 months ago

#259 - I train centerpoint with kitti, so how can i get rpn_centerhead_sim.onnx and centerpoint.scn.onnx??

Issue - State: open - Opened by ruishanyin 3 months ago

#258 - Jetson AGX Xavier with Jetpack 5.10 nvcc fatal : Unsupported gpu architecture 'compute_'

Issue - State: closed - Opened by HXB-1997 3 months ago - 1 comment

#257 - Dockerfile

Issue - State: open - Opened by Ankitzanzmera 3 months ago

#256 - Spconv Engine Implementation

Issue - State: open - Opened by barrydoooit 3 months ago - 4 comments

#255 - which model should I use?

Issue - State: open - Opened by mx2013713828 3 months ago - 5 comments

#254 - For the pretrained model bevfusion-det.pth, did you train it by youself, or download from other repository?

Issue - State: closed - Opened by OrangeJuice3211 3 months ago - 1 comment

#253 - How should I export the data of camera_bevfeat of type const nvtype::half*, because I want to visualize the BEV results of Camera. Thank you very much.

Issue - State: open - Opened by KaysenC 3 months ago - 2 comments

#252 - Failure in using Torch to execute PTQ checkpoint for BEVFusion inference

Issue - State: closed - Opened by barrydoooit 3 months ago - 2 comments

#251 - how to train and use myself data

Issue - State: open - Opened by xlmxbw 3 months ago - 1 comment

#250 - fix index of the last interval

Pull Request - State: closed - Opened by iceboundzw 4 months ago

#249 - why use "val = val.permute(4, 0, 1, 2, 3)" when export scn to onnx in centerpoint net

Issue - State: open - Opened by pangweisong 4 months ago - 1 comment

#248 - 有人转模型遇到过全是1的情况么

Issue - State: closed - Opened by zouzouwei 4 months ago - 1 comment

#247 - CenterPoint model

Issue - State: open - Opened by super-liuyang 4 months ago - 8 comments

#246 - lidar2image matrix problem

Issue - State: open - Opened by super-liuyang 4 months ago - 1 comment

#245 - How to specifiy a custom protobuf path instead of default path?

Issue - State: open - Opened by hitbuyi 5 months ago - 1 comment

#244 - Static version of `spconv` library?

Issue - State: open - Opened by josh-wende 6 months ago - 3 comments

#243 - export-scn.py Error:

Issue - State: open - Opened by Bonasdljlkj 6 months ago - 1 comment

#242 - How to modify the detection distance of the model？

Issue - State: open - Opened by zouzouwei 6 months ago - 4 comments

#241 - Python binding for cuOsd build failure

Issue - State: open - Opened by ksokolov-vaisto 6 months ago

#240 - Building the BEVFusion TRT engines via the C++ API instead of `trtexec`, they run but give garbage results

Issue - State: open - Opened by josh-wende 6 months ago - 2 comments

#239 - Half precision error when BEVFusion network built via C++ API instead of `trtexec`

Issue - State: closed - Opened by josh-wende 6 months ago - 2 comments

#238 - Building the BEVFusion TRT engines via C++ API, how to handle int8 calibration?

Issue - State: closed - Opened by josh-wende 6 months ago - 2 comments

#237 - Does the coordinate system affect the accuracy?

Issue - State: closed - Opened by san9569 6 months ago - 6 comments

#236 - CMake Compile Error

Issue - State: closed - Opened by JYS997760473 8 months ago - 4 comments

#235 - spconv/engine.hpp: No such file or directory

Issue - State: closed - Opened by k-nxt 8 months ago - 2 comments

#234 - cuda12 support for spconv v1.1.2

Issue - State: open - Opened by wep21 8 months ago - 12 comments
Labels: fixed

#233 - libspconv.so protobuf version

Issue - State: open - Opened by josh-wende 8 months ago - 3 comments

#232 - PTQ/QAT Does the centerpoint_qat_pipline.py code only do ptq/qat on the sparse backbone?

Issue - State: open - Opened by bigsheep2018 8 months ago

#231 - Modifying the parameters min_range, max_range, and voxel_size for voxelization results in a decrease in the accuracy of inference.

Issue - State: open - Opened by nainaigetuide 8 months ago - 4 comments

#230 - lidar intensity normalization

Issue - State: closed - Opened by josh-wende 9 months ago - 1 comment

#229 - Running time

Issue - State: open - Opened by super-liuyang 9 months ago - 8 comments

#228 - Reproduce the video of your demonstration

Issue - State: open - Opened by super-liuyang 9 months ago

#227 - How can I achieve the video in your demonstration by changes the configurations in the code

Issue - State: open - Opened by super-liuyang 9 months ago - 4 comments

#226 - BEVSegmentationHead

Issue - State: open - Opened by lucianzhong 9 months ago - 2 comments

#225 - cudaErrorInvalidConfiguration [ 9 ]

Issue - State: closed - Opened by wep21 9 months ago - 2 comments

#224 - Could you specify which exact model of the NVIDIA Jetson Orin series was used in this project?

Issue - State: open - Opened by catalysts-higher 10 months ago - 3 comments

#223 - fps result

Issue - State: open - Opened by Yunge6666 10 months ago - 3 comments

#222 - About camera-bevpool.cu and camera-geometry.cu

Issue - State: open - Opened by cn-hezhu 10 months ago - 2 comments

#221 - jetson 支持

Issue - State: closed - Opened by fengguibinyn 10 months ago - 1 comment

#220 - if the input size of the spconv onnx should be dynamic ?

Issue - State: open - Opened by pegasus-zhang 10 months ago - 1 comment

#219 - Doubts about “MaxDistance”

Issue - State: closed - Opened by Like2021 10 months ago - 5 comments

#218 - There is no resnet50 in bevfusion train code

Issue - State: closed - Opened by FantineFu 10 months ago - 5 comments

#217 - Different SparseConvolution results between C++ and pytorch

Issue - State: open - Opened by Vincentdaniele 10 months ago

#216 - function and parameters in lidar-voxelization.cu function and engine.hpp

Issue - State: closed - Opened by bujianyiwang 10 months ago - 4 comments

#215 - Deployment of lidar-only model

Issue - State: closed - Opened by imurfutur3 10 months ago - 2 comments

#214 - AttributeError: 'SparseConvolutionQunat' object has no attribute '_conv_forward'

Issue - State: open - Opened by wugoukanle 11 months ago - 1 comment

#211 - KeyError: ‘LoadRadarPointsMultiSweeps...’

Issue - State: open - Opened by Yunge6666 11 months ago - 2 comments

#209 - cuOSD stb font backend memory too big

Issue - State: open - Opened by zxpshadow 11 months ago - 5 comments
Labels: To be confirmed

#208 - Number of lidar sweeps used for CUDA-BEVFusion inference

Issue - State: closed - Opened by W6WM9M 11 months ago - 4 comments

#207 - pytorch_quantization/cuda_ext.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZN3c1015SmallVectorBaseIjE8grow_podEPvmm

Issue - State: open - Opened by Charmve 11 months ago - 4 comments

#190 - incorrect inference on KITTI dataset

Issue - State: closed - Opened by san9569 12 months ago - 2 comments

#182 - Assertion failed: scaleAllPositive && "Scale coefficients must all be positive" when build_trt_engine.sh

Issue - State: closed - Opened by san9569 about 1 year ago - 2 comments

#178 - intermediate sparse tensor not align

Issue - State: open - Opened by HuangVictorAuto about 1 year ago - 16 comments

#177 - Train and test on custom data.

Issue - State: open - Opened by krishnatoshniwal about 1 year ago - 2 comments

#175 - Inference time 25 FPS

Issue - State: open - Opened by dav695 about 1 year ago - 7 comments

#174 - error when replace load_images of loading ros msg

Issue - State: open - Opened by leslie27ch about 1 year ago - 6 comments

#166 - About voxelization parameters

Issue - State: open - Opened by catziyan about 1 year ago - 7 comments

#161 - About camera-geometry.cu

Issue - State: open - Opened by pangxiaowen about 1 year ago - 6 comments
Labels: To be confirmed

#159 - 你好，运行bash tool/run.sh报错

Issue - State: closed - Opened by xqh5201314 about 1 year ago - 4 comments

#151 - feature request for supporting for gpu arch sm75

Issue - State: open - Opened by HenryZhangJianhe about 1 year ago - 2 comments
Labels: Feature Request

#146 - libspconv.so build

Issue - State: open - Opened by skprot about 1 year ago - 15 comments
Labels: Feature Request

#135 - 'SwinTransformer' object has no attribute 'layer3'

Issue - State: open - Opened by kunnalparihar about 1 year ago - 10 comments

#130 - Does CUDA-BEVFusion support map segmentation model?

Issue - State: open - Opened by VanderBieu about 1 year ago - 4 comments

#128 - Can not find any matched kernel 9 x 32

Issue - State: open - Opened by akjt about 1 year ago - 8 comments
Labels: Feature Request

#112 - BEVFusion errors with different lidar configuration

Issue - State: closed - Opened by alre5639 about 1 year ago - 3 comments

#111 - The result of bevfusion int8 inference is incorrect

Issue - State: closed - Opened by Oh-Cest-la-vie about 1 year ago - 4 comments

#109 - result is incorrect

Issue - State: closed - Opened by MarioUncle-Wang about 1 year ago - 2 comments

#100 - Does libspconv support SparseInverseConv3d？

Issue - State: closed - Opened by CreepingBird over 1 year ago - 7 comments

#99 - Can you provide demo about onnx infer by TRT

Issue - State: closed - Opened by lix19937 over 1 year ago - 4 comments

#86 - Cmake Error

Issue - State: closed - Opened by alre5639 over 1 year ago - 12 comments

#85 - May I use my libspconv.so

Issue - State: closed - Opened by EmmanuelAntonioC over 1 year ago - 6 comments

GitHub / NVIDIA-AI-IOT/Lidar_AI_Solution issues and pull requests