[TensorRT-8] Add independent padding and depth NN #278

Merged
perseusdg merged 28 commits from tensorrt8 into tensorrt8 2022-03-30 15:42:34 +02:00

28 Commits

Author SHA1 Message Date
Harshvardhan Chandirasekar 40266a6c32 Fixed tensorrt8 branch to work jetpack 4.5 and tensorrt7
Signed-off-by: perseusdg <f20180523@goa.bits-pilani.ac.in>
2022-03-16 18:09:32 +05:30
perseusdg 480b5a9c5a added monodepth2_1024 2022-02-22 14:50:39 +00:00
perseusdg b3bc93693f Revert "Merge branch 'tensorrt8-rds' into depth"
This reverts commit 00f06f7bcc, reversing
changes made to 6a133d8dec.
2022-01-25 18:30:39 +05:30
Harshvardhan Chandirasekar ef564f134d Merge pull request #6 from perseusdg/depth
Merge - build fixes
2022-01-24 21:55:06 +05:30
perseusdg f996659845 Merge remote-tracking branch 'origin/depth' into depth 2022-01-24 21:48:35 +05:30
Harshvardhan Chandirasekar afdad8e661 Moved batchnorm and test_monodepth2_new_format layer to dev
Signed-off-by: perseusdg <harshvardhan.chandira@gmail.com>
2022-01-24 21:48:21 +05:30
Harshvardhan Chandirasekar 71366befc2 Merge pull request #5 from perseusdg/depth
Merge tensorrt8 and depth
2022-01-24 21:44:37 +05:30
Harshvardhan Chandirasekar 3e86671c50 Moved batchnorm and test_monodepth2_new_format layer to dev
Signed-off-by: perseusdg <harshvardhan.chandira@gmail.com>
2022-01-24 21:41:59 +05:30
Harshvardhan Chandirasekar 00f06f7bcc Merge branch 'tensorrt8-rds' into depth 2022-01-20 22:05:54 +05:30
Harshvardhan Chandirasekar 53725c88a9 Merge pull request #3 from perseusdg/tensorrt8
push tensorrt8 commits to the rds branch
2022-01-20 21:12:49 +05:30
Harshvardhan Chandirasekar 6a133d8dec -Replaced ISliceLayer based paddings(reflection,constant and zero) with the IPluginV2 version for TensorRT >= 8.2.0
-Fixed demo build issue on Windows

Signed-off-by: perseusdg <harshvardhan.chandira@gmail.com>
2022-01-20 00:52:22 +05:30
Micaela Verucchi decd73d298 Fix saving result video in demoDepth, add automatic download for monodepth2 weights
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
2022-01-19 10:22:06 -08:00
Micaela Verucchi 907df27e07 Fix monodepth2, add demoDepth:
- Fix monodepth2 network, now works with both cuDNN and tensorRT
- Substitute cuDNN ELU with tkDNN one
- add DepthNN class
- add demoDepth demo, now only works with monodepth2 net

Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
               Francesco Gatti <gattifrancesco@hotmail.it>
2022-01-18 21:34:16 -08:00
perseusdg 19e41a8b99 monodepth2 new format (test) conv + independent batch norm 2022-01-11 12:38:36 +05:30
perseusdg 061bc79a69 Added BatchNorm to NetworkRT (conver_layer) 2022-01-10 22:48:16 +05:30
perseusdg bcf0c4eab3 Added BatchNorm Layer (Testing still needs to be done) 2022-01-10 22:48:06 +05:30
perseusdg e1eac2d42a Added a seperate layer file for batchnorm,to support an independent batchnorm class in order to pass necessary parameters read from the bin file. 2022-01-10 22:47:59 +05:30
perseusdg 3023926695 fix constantpadding commit 2022-01-09 22:17:43 +05:30
perseusdg 4d8f99b441 depth->tensorrt8 patches 2022-01-09 19:24:38 +05:30
perseusdg cabebc95a2 added individual layer names in monodepth2 2022-01-09 13:45:44 +05:30
perseusdg 3b58fbcb82 Added ConstantPadding plugin for TensorRT < 8.2 2022-01-07 13:05:09 +05:30
perseusdg f189efcbb1 Reflection Padding native plugin fix ,forgot to added input_dim.c in the plugin creator 2022-01-07 05:41:26 +00:00
perseusdg 7298dcfb2f Added monodepth2.cpp 2022-01-06 16:30:11 +00:00
perseusdg b75cecb105 Added constant and zero padding with ISliceLayer,need to add them for tensorrt versions less than 8.2 using IPluginV2Ext instead of ISliceLayer since they dont seem to support reflect and zero 2022-01-04 09:03:04 +05:30
perseusdg ba022663f1 completed padding migrations from github 2022-01-03 19:49:32 +05:30
perseusdg 6837644eb2 add reflection padding from github 2022-01-03 18:58:13 +05:30
perseusdg dcf4054bc6 performance improvements for yolo based networks,significant reduction in inference time can be seen yolo4tiny ,yolo4 and minor reduction in inference time can be seen in yolo4_berkeley_f1 and yolo4_berkeley - tested with a batchsize of 1 and 2 and on gtx 1070,it is possible that the performance improvement is more signficant in newer hardware 2022-01-03 13:52:16 +05:30
perseusdg 19f12d9c15 rds_slam -> tensorrt8 port for rds 2021-11-15 00:08:37 +05:30