bob issueshttps://gitlab.idiap.ch/groups/bob/-/issues2017-10-18T16:36:36Zhttps://gitlab.idiap.ch/bob/bob.db.base/-/issues/19bob.db.base depends on bob.io.image without declaring that in the requirement...2017-10-18T16:36:36ZManuel Günthersiebenkopf@googlemail.combob.db.base depends on bob.io.image without declaring that in the requirements.txtWhen creating a database from scratch using `bob_db_manage.py`, I have the problem that `bob.io.image` is required, but it is not declared in the `requirements.txt`.
The reason is as follows:
1. While test cases are usually not part of ...When creating a database from scratch using `bob_db_manage.py`, I have the problem that `bob.io.image` is required, but it is not declared in the `requirements.txt`.
The reason is as follows:
1. While test cases are usually not part of the package, `setup.py` https://gitlab.idiap.ch/bob/bob.db.base/blob/15998f2dab6de6da677125c3eb19c42d61172c52/setup.py#L38 links the test database.
2. While `bob_db_manage.py` parses all databases, it tries to `import` the test database.
3. The test database requires `bob.io.image` to be present: https://gitlab.idiap.ch/bob/bob.db.base/blob/15998f2dab6de6da677125c3eb19c42d61172c52/bob/db/base/tests/sample/models.py#L14
* [ ] Quick fix: move the `import bob.io.image` into the function, where it is actually required.
* [ ] Real fix (includes the quick fix): add `bob.io.image` to the **test** requirementshttps://gitlab.idiap.ch/bob/bob.io.video/-/issues/7C++ bindings do not use the bob::extension documentation classes2017-10-19T14:01:12ZManuel Günthersiebenkopf@googlemail.comC++ bindings do not use the bob::extension documentation classesManuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.io.video/-/issues/5Reading audio from video files2017-10-19T14:01:12ZAmir MOHAMMADIReading audio from video filesIt would be great if we could read audio content from video files too.It would be great if we could read audio content from video files too.https://gitlab.idiap.ch/bob/bob.io.video/-/issues/4Unable to read a file because ffmpeg was not compiled with zlib support2017-10-19T14:01:12ZAndré AnjosUnable to read a file because ffmpeg was not compiled with zlib support*Created by: 183amir*
I think we need to add a test case and make sure that ffmpeg supports reading this file so that this does not happen again. The file is one of the Replay Attack database files.
```
$ ffprobe -i
Input #0, mov,mp...*Created by: 183amir*
I think we need to add a test case and make sure that ffmpeg supports reading this file so that this does not happen again. The file is one of the Replay Attack database files.
```
$ ffprobe -i
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'attack_highdef_client001_session01_highdef_photo_adverse.mov':
Metadata:
major_brand : qt
minor_version : 537199360
compatible_brands: qt
creation_time : 2010-10-14 09:17:26
Duration: 00:00:09.60, start: 0.000000, bitrate: 2584 kb/s
Stream #0:0(eng): Video: mjpeg (jpeg / 0x6765706A), yuvj420p(pc, bt470bg/unknown/unknown), 320x240 [SAR 72:72 DAR 4:3], 2583 kb/s, 25 fps, 25 tbr, 600 tbn, 600 tbc (default)
Metadata:
creation_time : 2010-10-14 09:17:26
handler_name : Apple Alias Data Handler
encoder : Photo - JPEG
```
```
$ mediainfo attack_highdef_client001_session01_highdef_photo_adverse.mov
General
Complete name : attack_highdef_client001_session01_highdef_photo_adverse.mov
Format : MPEG-4
Format profile : QuickTime
Format settings : Compressed header
Codec ID : qt 2005.03 (qt )
File size : 2.96 MiB
Duration : 9s 600ms
Overall bit rate mode : Constant
Overall bit rate : 2 585 Kbps
Encoded date : UTC 2010-10-14 09:17:26
Tagged date : UTC 2010-10-14 09:17:26
Writing library : Apple QuickTime
Video
ID : 1
Format : JPEG
Codec ID : jpeg
Duration : 9s 600ms
Bit rate mode : Constant
Bit rate : 2 584 Kbps
Width : 320 pixels
Height : 240 pixels
Display aspect ratio : 4:3
Frame rate mode : Constant
Frame rate : 25.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Compression mode : Lossy
Bits/(Pixel*Frame) : 1.346
Stream size : 2.96 MiB (100%)
Language : English
Encoded date : UTC 2010-09-16 07:25:56
Tagged date : UTC 2010-10-14 09:17:26
```
The output of ffmpeg without zlib support:
```
ffprobe -i attack_highdef_client001_session01_highdef_photo_adverse.mov
ffprobe version 2.8.6 Copyright (c) 2007-2016 the FFmpeg developers
built with gcc 4.4.7 (GCC) 20120313 (Red Hat 4.4.7-16)
configuration: --prefix=/opt/conda/envs/_build --disable-doc --disable-debug --disable-stripping --enable-avisynth --enable-avresample --enable-fontconfig --enable-gpl --enable-libfreetype --enable-libx264 --enable-pic --enable-shared --enable-static --extra-cxxflags=-DZLIB_CONST --enable-zlib
libavutil 54. 31.100 / 54. 31.100
libavcodec 56. 60.100 / 56. 60.100
libavformat 56. 40.101 / 56. 40.101
libavdevice 56. 4.100 / 56. 4.100
libavfilter 5. 40.101 / 5. 40.101
libavresample 2. 1. 0 / 2. 1. 0
libswscale 3. 1.101 / 3. 1.101
libswresample 1. 2.101 / 1. 2.101
libpostproc 53. 3.100 / 53. 3.100
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x2242ba0] this file requires zlib support compiled in
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x2242ba0] error reading header
```https://gitlab.idiap.ch/bob/bob.io.video/-/issues/2This package needs to be tested against ffmpeg 32017-10-19T14:01:12ZAndré AnjosThis package needs to be tested against ffmpeg 3*Created by: 183amir*
*Created by: 183amir*
https://gitlab.idiap.ch/bob/bob.io.video/-/issues/1Building Library twice2017-10-19T14:01:12ZAndré AnjosBuilding Library twice*Created by: anjos*
This package contains a `Library` definition inside its `setup.py` file that seems doubled (by the `Extension`). Is there a reason for this?*Created by: anjos*
This package contains a `Library` definition inside its `setup.py` file that seems doubled (by the `Extension`). Is there a reason for this?https://gitlab.idiap.ch/bob/bob.db.mnist/-/issues/4Move M-NIST files to our metafile server2017-10-19T14:04:35ZAndré AnjosMove M-NIST files to our metafile serverI think we should just package this ourselves and be done. We get issues at every build trying to download these files from some place else.I think we should just package this ourselves and be done. We get issues at every build trying to download these files from some place else.https://gitlab.idiap.ch/bob/bob.db.mnist/-/issues/3db.query(labels=0) returns all data2017-10-19T14:04:35ZAndré Anjosdb.query(labels=0) returns all data*Created by: siebenkopf*
When I query the database with:
>>> db = xbob.db.mnist.Database()
>>> training_targets = db.query("training", labels=0)[1]
>>> print (training_targets)
I get all possible targets, rather than only the target...*Created by: siebenkopf*
When I query the database with:
>>> db = xbob.db.mnist.Database()
>>> training_targets = db.query("training", labels=0)[1]
>>> print (training_targets)
I get all possible targets, rather than only the targets with label 0.
For all other labels, this work as expected.
I guess the problem is that the value ``0`` is special since ``0`` evaluates to ``False`` when tested, e.g., using ``if``. I will investigate.https://gitlab.idiap.ch/bob/bob.db.mnist/-/issues/2Package downloads M-NIST digit rec. DB at every iteration2017-10-19T14:04:35ZAndré AnjosPackage downloads M-NIST digit rec. DB at every iteration*Created by: anjos*
This package is downloading the M-NIST database from the respective URL (LeCun's site) at every iteration. The reason is that it has no way to know where it previously downloaded a copy (because it is using a tempora...*Created by: anjos*
This package is downloading the M-NIST database from the respective URL (LeCun's site) at every iteration. The reason is that it has no way to know where it previously downloaded a copy (because it is using a temporary directory to do so).
It would be more useful if the database was downloaded to a known directory.
I'd propose this database provided a "create" action that would allow it to be downloaded and installed within the package.https://gitlab.idiap.ch/bob/bob.db.mnist/-/issues/1Database is not (yet) part of the xbob.db.aggregator2017-10-19T14:04:35ZAndré AnjosDatabase is not (yet) part of the xbob.db.aggregator*Created by: laurentes*
We might consider to add it later on to run the test suite every night.*Created by: laurentes*
We might consider to add it later on to run the test suite every night.https://gitlab.idiap.ch/bob/bob.learn.boosting/-/issues/1several "assert" statements in the code2017-10-20T02:22:54ZAndré Anjosseveral "assert" statements in the code*Created by: siebenkopf*
it appears that the C++ code contains several ``assert`` statements, which will either be ignored, i.e., when compiled with ``-DNDEBUG`` (the default!!) or raise C-Style exceptions, which cannot be caught by the...*Created by: siebenkopf*
it appears that the C++ code contains several ``assert`` statements, which will either be ignored, i.e., when compiled with ``-DNDEBUG`` (the default!!) or raise C-Style exceptions, which cannot be caught by the Python bindings.
Is it possible to change these asserts into proper tests that raise exceptions?https://gitlab.idiap.ch/bob/bob.bio.face/-/issues/24Contradicting licenses: GPL and BSD2017-10-20T02:30:35ZSébastien MARCELContradicting licenses: GPL and BSD@tiago.pereira @amohammadi @mguenther @andre.anjos the files COPYING and LICENCE are contradicting -- COPYING contains a GPL clause and LICENCE is BSD-like
it should be fixed since I think we agreed that this package should be BSD@tiago.pereira @amohammadi @mguenther @andre.anjos the files COPYING and LICENCE are contradicting -- COPYING contains a GPL clause and LICENCE is BSD-like
it should be fixed since I think we agreed that this package should be BSDhttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/23`Algorithm.read_probe` should be removed2017-10-20T02:30:35ZManuel Günthersiebenkopf@googlemail.com`Algorithm.read_probe` should be removedAny `read_probe` function of any algorithm is deprecated and no longer used, after bob/bob.bio.base!78 is merged. It would be good to remove them.
This includes changing the according test cases, which need to be updated before bob/bob...Any `read_probe` function of any algorithm is deprecated and no longer used, after bob/bob.bio.base!78 is merged. It would be good to remove them.
This includes changing the according test cases, which need to be updated before bob/bob.bio.base!78 can be merged.Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/21baselines script cannot work if bob.bio.face is pip installed2017-10-20T02:30:35ZAndré Anjosbaselines script cannot work if bob.bio.face is pip installedThe reason is that it has hard-coded paths to `./bin/verify.py` and `./bin/evaluate.py`.The reason is that it has hard-coded paths to `./bin/verify.py` and `./bin/evaluate.py`.Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/19Bug in LFW extension2017-10-20T02:30:35ZTiago de Freitas PereiraBug in LFW extensionThe scores of an experiment using LFW gives me this output, which is obviously incorrect:
```
13 Enos_Slaughter Enos_Slaughter/Enos_Slaughter_0001 -0.32567173
13 Juan_Carlos_Ortega Juan_Carlos_Ortega/Juan_Carlos_Ortega_0001 -0.33916610
...The scores of an experiment using LFW gives me this output, which is obviously incorrect:
```
13 Enos_Slaughter Enos_Slaughter/Enos_Slaughter_0001 -0.32567173
13 Juan_Carlos_Ortega Juan_Carlos_Ortega/Juan_Carlos_Ortega_0001 -0.33916610
13 Marlon_Devonish Marlon_Devonish/Marlon_Devonish_0001 -0.31824485
18 Abdel_Nasser_Assidi Abdel_Nasser_Assidi/Abdel_Nasser_Assidi_0002 -0.04248074
51 Bruce_Paltrow Bruce_Paltrow/Bruce_Paltrow_0001 -0.40190915
51 Howard_Smith Howard_Smith/Howard_Smith_0002 -0.37437308
51 Jan_De_Bont Jan_De_Bont/Jan_De_Bont_0001 -0.41860976
51 Jim_Nochols Jim_Nochols/Jim_Nochols_0001 -0.27871283
51 Oleg_Romantsev Oleg_Romantsev/Oleg_Romantsev_0001 -0.35602057
```
For most of the databases the `model_id` is equals to `client_id`, which is not the case for this dataset.
`bob.bio.face.database.LFWBioDatabase` should reimplement the method `bob.bio.face.database.BioDatabase.client_id_from_model_id` and fix this.Tiago de Freitas PereiraTiago de Freitas Pereirahttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/18The cuhk_cufs is not tested2017-10-20T02:30:35ZAmir MOHAMMADIThe cuhk_cufs is not testedsince there is no config file for it.since there is no config file for it.Refactoring 2016 and gitlab migration milestoneTiago de Freitas PereiraTiago de Freitas Pereirahttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/17This package documentation depends on itself2017-10-20T02:30:35ZAndré AnjosThis package documentation depends on itselfThis one is great: this package documentation depends on itself to build ;-)
Please fix this ASAP.This one is great: this package documentation depends on itself to build ;-)
Please fix this ASAP.Tiago de Freitas PereiraTiago de Freitas Pereirahttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/16dlib functions2017-10-20T02:30:35ZTiago de Freitas Pereiradlib functionsHi,
As some may know, dlib toolkit has some very nice features for face based biometrics.
It has a :
- Very good landmark localization
- Pose estimation
- Face detection
- ......
Would be nice to have this in `bob.bio.face` pr...Hi,
As some may know, dlib toolkit has some very nice features for face based biometrics.
It has a :
- Very good landmark localization
- Pose estimation
- Face detection
- ......
Would be nice to have this in `bob.bio.face` pre-processors.
They have a package on conda-forge, so for us would be straight forward to integrate.
We can integrate it either directly on `bob.bio.face` or in another package `bob.bio.face_dlib`.
What do you think?Tiago de Freitas PereiraTiago de Freitas Pereirahttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/15IJB-A database implemented incorrectly2017-10-20T02:30:35ZManuel Günthersiebenkopf@googlemail.comIJB-A database implemented incorrectlyThe IJB-A database interface is not implemented correctly, and it is not tested -- the tests are skipped with an error message (see e.g. https://gitlab.idiap.ch/bob/bob.bio.face/builds/26262):
```
bob.bio.face.test.test_databases.test_i...The IJB-A database interface is not implemented correctly, and it is not tested -- the tests are skipped with an error message (see e.g. https://gitlab.idiap.ch/bob/bob.bio.face/builds/26262):
```
bob.bio.face.test.test_databases.test_ijba ... SKIP: Skipping test since the database bob.db.ijba seems not to be available: The given command line option 'ijba' is neither a resource for a 'database', nor an existing configuration file, nor could be interpreted as a command (error: name 'ijba' is not defined)
```
IJB-A needs to implement the BioFileSet protocol, but in its current version, this is not done. Also, for IJB-A we do not have all annotations for all images, so the test here: https://gitlab.idiap.ch/bob/bob.bio.face/blob/master/bob/bio/face/test/test_databases.py#L123 will most probably fail.
Also, we have the same issue for the annotations as mentioned in #14 and https://gitlab.idiap.ch/bob/bob/issues/235.
I can provide an implementation for the database, making sure that all the issues are solved. We might need to change the implementation of the annotations, when we have come up with a generic solution for #14.Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/14Some database interfaces do not provide access to the annotations stored in t...2017-10-20T02:30:35ZManuel Günthersiebenkopf@googlemail.comSome database interfaces do not provide access to the annotations stored in the low-level databasesAs discussed in https://gitlab.idiap.ch/bob/bob/issues/235, there are some databases that do not take care of the annotations stored in the low-level databases. New `BioDatabase` interfaces (and new `BioFile` interfaces) need to be provi...As discussed in https://gitlab.idiap.ch/bob/bob/issues/235, there are some databases that do not take care of the annotations stored in the low-level databases. New `BioDatabase` interfaces (and new `BioFile` interfaces) need to be provided.
I think, having a generic `BioFile` derivative (maybe called `BioFileFromLowLevel` or alike) that keeps track of the low-level `File` object (which is required to query the annotations) would be easier than writing specialized `XXXBioFile`'s for each case. Also, a generic `BioDatabaseWithAnnotations` can be created, from which the `XXXBioDatabase`'s can be derived, so that the implementation needs to be done only once.
The question is A) if we want to have these intermediate level `BioFileFromLowLevel` and `BioDatabaseWithAnnotations` implementations and B) where we should put them? Is there any need for such intermediate level classes in other packages than `bob.bio.face` (and `bob.bio.video`)?Refactoring 2016 and gitlab migration milestone