1
0
mirror of https://github.com/opencv/opencv_contrib.git synced 2025-10-25 04:26:17 +08:00
Commit Graph

32 Commits

Author SHA1 Message Date
Vladimir
2858b7bbad Fixed segmentation fault error in TRACK_vot::getNextFrame() 2015-08-28 01:30:18 +09:00
Vladimir
4e1d8ed8b9 Fixed argument dataset index in examples 2015-08-28 01:30:17 +09:00
Vladimir
6b3413339f Fix: - Get dataset path form *args 2015-08-28 01:17:41 +09:00
Vladimir
071092d97c Added VOT2015 Dataset interface + VOT2015 sample cpp 2015-08-28 01:17:40 +09:00
Vladimir
20cdf5d7f6 Fixed segmentation fault error in TRACK_vot::getNextFrame() 2015-08-28 01:12:13 +09:00
Vladimir
a4fdf7efff Fixed argument dataset index in examples 2015-08-28 01:11:29 +09:00
Vladimir
7b1fd98e15 Fix: - Get dataset path form *args 2015-08-28 01:08:40 +09:00
Vladimir
769b28e09c Added VOT2015 Dataset interface + VOT2015 sample cpp 2015-08-28 01:08:39 +09:00
Vadim Pisarevsky
60f5e5ef80 Merge pull request #327 from lluisgomez:svt_dataset 2015-08-24 17:00:43 +00:00
Vadim Pisarevsky
f9dab6164c Merge pull request #328 from lluisgomez:icdar_dataset 2015-08-24 17:00:20 +00:00
Lluis Gomez-Bigorda
45f4bd9828 Add benchmark code for the Chars74k dataset. Using the CNN character classifier reaches 75% and 84% accuracy for case-sensitive and case-insensitive recognition respectively. 2015-08-21 18:54:52 +02:00
lluis
271a9b4091 Fix w64 warnings 2015-08-09 08:51:08 +02:00
lluis
6f4a09ca7e Add benchmark for ICDAR2015 dataset using OCRTesseract and ERFilter classes. Gives word spotting f-score 0.642082 with strongly contextualized lexicon (100 words). 2015-08-09 00:13:40 +02:00
lluis
be927d962a Report the total f1 score (0.37) and not the mean-f1. This is in accordance with the standard SVT evaluation protocol and allows for comparison with other published results. 2015-08-06 11:59:53 +02:00
lluis
769d2ada0e Changes SVT recognition evaluation to be Case Insensitive (according to the standard evaluation protocol). This makes the benchmark obtained mean-f1 score increase from 0.23 to 0.27 2015-08-06 11:53:46 +02:00
Lluis Gomez-Bigorda
306aed4078 Adds interface for the ICDAR2015 dataset (end-to-end text recognition task) 2015-08-06 10:54:58 +02:00
Vadim Pisarevsky
7f4c627ec7 Merge pull request #240 from Jiaolong/dataset_pascal
added PASCAL VOC dataset
2015-07-01 12:20:51 +03:00
jiaolong_x220
8fd8703ce8 added PASCAL VOC dataset 2015-06-29 09:11:22 +02:00
jiaolong_x220
f7a5bab785 added INRIA pedestrian dataset 2015-06-12 04:29:41 +02:00
Vadim Pisarevsky
bcb5c6d7c3 Merge pull request #176 from berak/lfw_benchmark
add a 'split' train method to datasets/lfw_benchmark
2015-04-03 14:59:41 +03:00
berak
04dfaaf935 add a 'split' train method to datasets/lfw_benckmark 2015-02-20 14:55:26 +01:00
Maksim Shabunin
34131405df Removed some methods from cv::Algorithm and changed ml module interfaces 2015-02-17 11:45:57 +03:00
Dmitriy Anisimov
f04f191111 minor update of ar_hmdb code 2014-11-15 13:43:59 +03:00
Dmitriy Anisimov
3929e23878 updated adience loader 2014-11-15 09:08:52 +03:00
Dmitriy Anisimov
38168334f6 added first version of hpe_humaneva dataset loader 2014-10-30 13:12:18 +03:00
Dmitriy Anisimov
44f47427df updated or_imagenet dataset loader 2014-10-22 16:16:54 +04:00
Dmitriy Anisimov
3be484f12b added tr_svt benchmark 2014-10-21 13:21:35 +04:00
Dmitriy Anisimov
c5bc6e82ea added fr_adience dataset loader 2014-10-21 12:49:46 +04:00
Dmitriy Anisimov
e16ba0430a updated or_sun dataset loader 2014-10-18 16:20:26 +04:00
dmitriy.anisimov
2c8c4ddf77 added benchmark for ar_hmdb 2014-10-10 14:19:37 +04:00
Dmitriy Anisimov
a5351d9e86 added Caltech Pedestrian dataset loading 2014-10-10 12:43:37 +04:00
Dmitriy Anisimov
74fbe96b6f renamed datasetstools to datasets 2014-09-23 14:08:09 +04:00