Commit Graph

9 Commits

Author SHA1 Message Date
Wang Xinyu
cb7c8a07d4 [CVCUDA] PaddleDetection preprocessor support CV-CUDA (#1493)
* ppdet preproc use manager

* pad_to_size chw opencv

* pad_to_size chw flycv

* fix pad_to_size flycv

* add warning message

* cvcuda convert cubic to linear, padToSize cvcuda

* stridedpad cvcuda

* fix flycv include

* fix flycv include

* fix flycv build

* cast cvcuda

* fix pybind

* fix normalize permute cuda

* base processor move funcs to cc

* Update pad_to_size.cc
2023-03-10 12:43:57 +08:00
Wang Xinyu
91a1c72f98 [CVCUDA] PP-OCR detector preprocessor integrate CV-CUDA (#1382)
* move manager initialized_ flag to ppcls

* update dbdetector preprocess api

* declare processor op

* ppocr detector preprocessor support cvcuda

* move cvcuda op to class member

* ppcls use manager register api

* refactor det preprocessor init api

* add set preprocessor api

* add create processor macro

* new processor call api

* ppcls preprocessor init resize on cpu

* ppocr detector preprocessor set normalize api

* revert ppcls pybind

* remove dbdetector set preprocessor

* refine dbdetector preprocessor includes

* remove mean std in py constructor

* add comments

* update comment

* Update __init__.py
2023-02-22 19:39:11 +08:00
Wang Xinyu
efa46563f3 [nvJPEG] Integrate nvJPEG decoder (#1288)
* nvjpeg cmake

* add common decoder, nvjpeg decoder and add image name predict api

* ppclas support nvjpeg decoder

* remove useless comments

* image decoder support opencv

* nvjpeg decode fallback to opencv

* fdtensor add nbytes_allocated

* single image decode api

* fix bug

* add pybind

* ignore nvjpeg on jetson

* fix cmake in

* predict on fdmat

* remove image names predict api, add image decoder tutorial

* Update __init__.py

* fix pybind
2023-02-17 10:27:05 +08:00
Wang Xinyu
d3d914856d [CVCUDA] Utilize CV-CUDA batch processing function (#1223)
* norm and permute batch processing

* move cache to mat, batch processors

* get batched tensor logic, resize on cpu logic

* fix cpu compile error

* remove vector mat api

* nits

* add comments

* nits

* fix batch size

* move initial resize on cpu option to use_cuda api

* fix pybind

* processor manager pybind

* rename mat and matbatch

* move initial resize on cpu to ppcls preprocessor

---------

Co-authored-by: Jason <jiangjiajun@baidu.com>
2023-02-07 13:44:30 +08:00
heliqi
1427d5d29a [Bug] Vision and text compile source file add .cu file on CMakeLists.txt (#1188)
* vision and text CMake add .cu file

* recover .cu file

* recover .cu file

* vision and text CMake add .cu file

* vision and text CMake add .cu file

* vision and text CMake add .cu file

* Update yolo_preprocess.cu
2023-01-31 11:39:12 +08:00
Jason
774e4d26fd [Bug] Fix compile problem (#1186)
Fix compile problem
2023-01-30 14:25:01 +08:00
Wang Xinyu
62e051e21d [CVCUDA] CMake integration, vison processor CV-CUDA integration, PaddleClas support CV-CUDA (#1074)
* cvcuda resize

* cvcuda center crop

* cvcuda resize

* add a fdtensor in fdmat

* get cv mat and get tensor support gpu

* paddleclas cvcuda preprocessor

* fix compile err

* fix windows compile error

* rename reused to cached

* address comment

* remove debug code

* add comment

* add manager run

* use cuda and cuda used

* use cv cuda doc

* address comment

---------

Co-authored-by: Jason <jiangjiajun@baidu.com>
2023-01-30 09:33:49 +08:00
Jason
054dfec43a [Other] Remove some build options (#1090)
* remove some flags

* add gpu check in cmake
2023-01-09 17:36:07 +08:00
Wang Xinyu
a36f5d3396 [Backend] cuda normalize and permute, cuda concat, optimized ppcls, ppdet & ppseg (#546)
* cuda normalize and permute, cuda concat

* add use cuda option for preprocessor

* ppyoloe use cuda normalize

* ppseg use cuda normalize

* add proclib cuda in processor base

* ppcls add use cuda preprocess api

* ppcls preprocessor set gpu id

* fix pybind

* refine ppcls preprocessing use gpu logic

* fdtensor device id is -1 by default

* refine assert message

Co-authored-by: heliqi <1101791222@qq.com>
2022-11-14 18:44:00 +08:00