vato007/fast-depth-tf
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 950dd4582e Start adding optimise step to unsupervised train, add demo to main unsupervised Piv 2022-05-03 19:51:34 +09:30
  • 909c38cf56 Fix total loss calculation, add some more todos Piv 2022-05-03 19:05:29 +09:30
  • 188c55d1c8 Update for python 2.10, add general training algorithm step piv 2022-05-03 16:56:15 +09:30
  • aa423cc38a Start adding unsupervised train loop Piv 2021-11-20 13:37:26 +10:30
  • 513227a84a Merge branch 'coreml-2' into 'main' main Michael Pivato 2021-09-13 23:43:48 +00:00
  • a053238310 Fix dense depth loss and model, fix metric naming Piv 2021-09-13 20:35:59 +09:30
  • 28b11aaa44 Use image type for CoreML input, reshape last layer to work with metal shader Piv 2021-09-10 22:13:41 +09:30
  • 2bb37b2722 Fix up generator to include intrinsics Piv 2021-08-29 19:26:15 +09:30
  • 90b73bf420 Start adding generators for unsupervised training Piv 2021-08-29 18:06:37 +09:30
  • 42fcf5554a Update warp test to verify output shape Piv 2021-08-24 21:39:10 +09:30
  • c164c9720a Finish Projective Inverse Warp algorithm Piv 2021-08-24 20:13:30 +09:30
  • b7917ec465 More warp implementation Piv 2021-08-21 17:32:16 +09:30
  • df1ac89a81 Add euler to rotation matrix, grid flattening Piv 2021-08-10 20:39:52 +09:30
  • 8016f0f945 Add coordinates generation implementation Piv 2021-08-08 22:11:50 +09:30
  • ece37843ce Merge branch 'main' into unsupervised Piv 2021-08-08 18:58:39 +09:30
  • a72f04006f Merge branch 'packnet_small' into 'main' Michael Pivato 2021-08-08 09:25:56 +00:00
  • 625ecba731 Add small option to packnet, fix docs and first/final conv layers (prev 32) Michael Pivato 2021-08-08 09:25:55 +00:00
  • 5373dc6b65 Merge branch 'packnet_fixes' into 'main' Michael Pivato 2021-08-07 11:31:22 +00:00
  • 58b8e53986 Fix packnet residual block and layers, refactor to support different amount of residual layers Piv 2021-08-07 21:00:17 +09:30
  • cd278e683f Start adding pose warp conversions Piv 2021-08-07 17:18:06 +09:30
  • f56e663fca Merge branch 'unsupervised' into 'main' Michael Pivato 2021-08-05 08:20:31 +00:00
  • 26dda68523 Add Smooth Loss Michael Pivato 2021-08-05 08:20:31 +00:00
  • 5996d6eaf0 Merge branch 'main' into unsupervised Piv 2021-08-05 17:49:48 +09:30
  • 8be4ce4e6d Add smooth loss Piv 2021-08-05 17:48:44 +09:30
  • e96e6c2c2b Merge branch 'unsupervised' into 'main' Michael Pivato 2021-08-04 11:36:30 +00:00
  • 5d0731b60f PoseNet Michael Pivato 2021-08-04 11:36:30 +00:00
  • b95442bb23 Finish off pose net Piv 2021-08-04 20:51:46 +09:30
  • a111f89722 Start adding pose decoder Piv 2021-08-03 20:25:19 +09:30
  • 2372b906df Add resnet18 Piv 2021-08-01 10:44:33 +09:30
  • d3a63c6bcd Merge branch 'unsupervised' into 'main' Michael Pivato 2021-07-31 01:31:25 +00:00
  • 6514fb0e86 Add sample openvino inference Piv 2021-07-28 20:15:15 +09:30
  • 3254eef4bf Add compiling packnet model, refactor modules to not duplicate loaders and trainers Piv 2021-07-23 22:41:46 +09:30
  • 66cbc7faf6 Add pack layer tests, fix unpack_3d layer Piv 2021-07-19 20:37:54 +09:30
  • 38e7ad069e Refactor load/util, start fixing packnet to support NHWC format Piv 2021-07-19 12:32:56 +09:30
  • d8bf493999 More packnet implementation Piv 2021-07-18 19:54:28 +09:30
  • 603de2bc9f Start adding packnet model Piv 2021-07-18 18:59:25 +09:30
  • e372fe33ba Add per-pixel loss functions Piv 2021-07-13 20:32:45 +09:30
  • b9457f17fe Add pixel loss functions, move warp Piv 2021-07-13 19:20:24 +09:30
  • 6f7da21977 Add spatial transformer network sampler Piv 2021-07-12 18:19:59 +09:30
  • f501beb6f2 Start implementing unsupervised train loop, add sfmlearner train and utils files for reference Piv 2021-07-05 20:50:12 +09:30
  • ba0ba609a3 Add utils from sfm learner, start adding sfm learner loss function to work with keras Piv 2021-06-22 22:03:23 +09:30
  • 101fe08924 Start adding stubs for unsupervised training Piv 2021-06-16 21:55:13 +09:30
  • e5b07fb766 Fix loss function Piv 2021-06-16 21:52:31 +09:30
  • 0547509689 Merge branch 'kitti_depth_dataset' into 'main' Michael Pivato 2021-04-22 12:13:48 +00:00
  • 070aec6eed Add Kitti depth dataset Michael Pivato 2021-04-22 12:13:48 +00:00
  • 02d8cd5810 Remove half-features from dense_depth Piv 2021-04-22 12:30:30 +09:30
  • acdb58396c Remove usage of keras Piv 2021-04-14 12:45:01 +09:30
  • cf7d2561ec Implement details of dense depth paper Piv 2021-04-14 12:38:51 +09:30
  • f598005b73 Add basic coreml and mlkit conversion scripts Piv 2021-03-29 18:58:16 +10:30
  • f3fc0f8fbb Add coreml conversion Piv 2021-03-29 18:58:16 +10:30
  • f2a42cca4c Merge branch 'dense-depth' into 'main' Michael Pivato 2021-03-29 07:31:42 +00:00
  • d88e9d3f12 Add dense-depth and experimental dense-net-nnconv5 models Piv 2021-03-29 17:59:18 +10:30
  • 870429c3ef Refactor fast-depth Piv 2021-03-29 17:55:07 +10:30
  • 3325ea0c0c Format pep8, include pass shaped to mobilenet Piv 2021-03-25 21:56:50 +10:30
  • 9449ddef01 Add notebook, gitignore Piv 2021-03-25 21:50:19 +10:30
  • 78d5aace15 Remove Experimental model Piv 2021-03-25 21:28:07 +10:30
  • ab7da5acd4 Add documentation and README, use Upsampling2D rather than image Resizing layer Piv 2021-03-24 21:35:25 +10:30
  • ac3ab27ddd Add model with no batch normalisation, use actual mobilnet model rather than extracting layers Piv 2021-03-24 20:00:26 +10:30
  • 39074f22a7 Add working train and eval functions for nyu_v2 Piv 2021-03-21 09:51:45 +10:30
  • fea08521bb Add metrics, prepare for training Piv 2021-03-17 21:14:38 +10:30
  • 00762f3e86 Build Functional Model, remove subclassed model Piv 2021-03-17 18:24:46 +10:30
  • b25b9be4eb Initial Commit Piv 2021-03-16 21:06:27 +10:30