Commits

Breadcrumbs

History for

onmaster

User selector

All users

All time

Commit History

Commits on Jan 5, 2018

• Fixed missing #include "../precomp.hpp"

  

  arthur-williams

  authored and

  alalek

  committed
  8a67858

  Copy full SHA for 8a67858

Commits on Jan 4, 2018

• add ocl implementation of proposal layer

  Show description for 34bfd7e
  pli2-intel

  committed
  34bfd7e

  Copy full SHA for 34bfd7e

Commits on Dec 29, 2017

• cmake(opt): AVX512_SKX

  alalek

  committed
  7d67d60

  Copy full SHA for 7d67d60

Commits on Dec 28, 2017

• Merge pull request #10456 from dkurt:dnn_allocate_mem_for_optimized_concat

  alalek

  committed
  8e7af7f

  Copy full SHA for 8e7af7f

• Merge pull request #10416 from fenrus75:avx512

  alalek

  committed
  a65b5df

  Copy full SHA for a65b5df

• cmake: AVX512 -> AVX_512F

  alalek

  committed
  898ca38

  Copy full SHA for 898ca38

• Allocate new memory for optimized concat to prevent collisions.

  Show description for a9807d8
  dkurt

  committed
  a9807d8

  Copy full SHA for a9807d8

• Add ocl version FasterRCNN accuracy test

  Show description for 00f03c5
  pli2-intel

  committed
  00f03c5

  Copy full SHA for 00f03c5

• Merge pull request #10424 from dkurt:fix_concat_optim

  alalek

  committed
  99a9c10

  Copy full SHA for 99a9c10

Commits on Dec 27, 2017

• Merge pull request #10428 from pengli:dnn

  alalek

  committed
  f3880c6

  Copy full SHA for f3880c6

Commits on Dec 26, 2017

• Provide a few AVX512 optimized functions for the DNN module

  Show description for 2938860
  fenrus75

  committed
  2938860

  Copy full SHA for 2938860

• Limit Concat layer optimization

  dkurt

  committed
  70c605a

  Copy full SHA for 70c605a

• dnn(ocl4dnn): update pre-tuned kernel config

  Show description for 84e2fa7
  pli2-intel

  committed
  84e2fa7

  Copy full SHA for 84e2fa7

Commits on Dec 23, 2017

• build: fix MSVS2010 build error

  alalek

  committed
  adf43e7

  Copy full SHA for adf43e7

Commits on Dec 22, 2017

• Merge pull request #10402 from dkurt:dnn_tf_quantized

  alalek

  committed
  019b7c5

  Copy full SHA for 019b7c5

• Merge pull request #10385 from pengli:dnn

  alalek

  committed
  59e825e

  Copy full SHA for 59e825e

• TensorFlow weights dequantization

  dkurt

  committed
  bcc669f

  Copy full SHA for bcc669f

• Merge pull request #10397 from mshabunin:fix-incorrect-assert

  alalek

  committed
  97af608

  Copy full SHA for 97af608

• add one more convolution kernel tuning candidate

  Show description for 181b448
  pli2-intel

  committed
  181b448

  Copy full SHA for 181b448

• Merge pull request #10265 from dkurt:nms_for_region_layer

  vpisarev

  committed
  0742e12

  Copy full SHA for 0742e12

• Replaced incorrect CV_Assert calls with CV_Error

  mshabunin

  committed
  aa46e31

  Copy full SHA for aa46e31

• Merge pull request #10364 from dkurt:dnn_smooth_tf_data_layout

  vpisarev

  committed
  325cbd7

  Copy full SHA for 325cbd7

Commits on Dec 21, 2017

• cleanup unnecessary macros in convolution ocl kernel

  Show description for c5fc8e0
  pli2-intel

  committed
  c5fc8e0

  Copy full SHA for c5fc8e0

• refactor candidate generation of convolution auto-tuning

  Show description for 0aa5e43
  pli2-intel

  committed
  0aa5e43

  Copy full SHA for 0aa5e43

• Refactor NMS procedure at RegionLayer

  dkurt

  committed
  c67e75b

  Copy full SHA for c67e75b

Commits on Dec 20, 2017

• Manage TensorFlow's NHWC data layout is smoother

  dkurt

  committed
  7e48fa5

  Copy full SHA for 7e48fa5

• R-FCN models support

  dkurt

  committed
  0ed2cbc

  Copy full SHA for 0ed2cbc

Commits on Dec 19, 2017

• Merge pull request #10341 from pengli:dnn

  alalek

  committed
  dcdd6af

  Copy full SHA for dcdd6af

• add depthwise convolution kernel

  Show description for 436d7e4
  pli2-intel

  committed
  436d7e4

  Copy full SHA for 436d7e4

• prior box layer ocl implementation

  Show description for 910d7da
  pli2-intel

  committed
  910d7da

  Copy full SHA for 910d7da

Commits on Dec 18, 2017

• Remove cv::dnn::Importer

  dkurt

  committed
  6aabd6c

  Copy full SHA for 6aabd6c

Commits on Dec 17, 2017

• Fix default pooling layer type

  dkurt

  committed
  2b43d4f

  Copy full SHA for 2b43d4f

Commits on Dec 15, 2017

• Fixed 3 issues found by static analysis

  mshabunin

  committed
  1033f2b

  Copy full SHA for 1033f2b

• Merge pull request #10306 from dkurt:faster_rcnn

  vpisarev

  committed
  62359f7

  Copy full SHA for 62359f7

• Faster-RCNN models support

  dkurt

  committed
  08112f3

  Copy full SHA for 08112f3

Pagination

Previous

Next