profile
viewpoint
Ignacio Arganda-Carreras iarganda Ikerbasque / Basque Country University (UPV/EHU) Donostia-San Sebastian, Spain https://sites.google.com/site/iargandacarreras/

AdrianNunez/Fall-Detection-with-CNNs-and-Optical-Flow 126

Repository containing the material required to reproduce the results of the paper "Vision-Based Fall Detection with Convolutional Neural Networks"

Borda/BIRL 51

BIRL: Benchmark on Image Registration methods with Landmark validations

danifranco/EM_Image_Segmentation 3

Deep Learning approaches for EM data

hornm/imagej-pixml 2

Pixel-based Machine Learning for ImageJ and beyond

iarganda/watershed-3d 2

ImageJ/Fiji library and plugin to perform watershed in 3D images

Vicomtech/serverless-mlperf 2

This repo aims to benchmark Amazon AWS DNN performance with Caffe, TensorFlow and OpenVINO models, using OpenCV and OpenVINO IE as inference backend engines.

CVPD/Trainable_Superpixel_Segmentation 1

This plugin has as a purpose the supervised classification of images using superpixels. To do so this plugin uses ImageJ, Weka and MorphoLibJ. ImageJ provides image handling support, Weka supports classification support and MorphoLibJ provides image feature extraction based on superpixel images.

esgomezm/NEUBIAS_chapter_DL_2020 0

Tthis repository contains all the material for the chapter "Building a Bioimage Analysis Workflow using Deep Learning"

push eventijpb/MorphoLibJ

dlegland

commit sha 64605c1d4ca97fcd597e1225a5bd5d22ace6d0f0

fix computation of max values when max value is negative

view details

push time in 5 minutes

push eventdanifranco/EM_Image_Segmentation

Daniel Franco

commit sha ffd1faa4d131fffeed66d6676083b2ff9eb14990

Change Oztel paper link

view details

push time in 38 minutes

push eventdanifranco/EM_Image_Segmentation

Daniel Franco

commit sha 429afa036441bd441b2292ed41fafd948e359287

Adapt Oztel code to doc

view details

push time in an hour

push eventdanifranco/EM_Image_Segmentation

Daniel Franco

commit sha c70950d6534d2a293784a129edbeb7a1eea00cda

Delete LRFinder dependency

view details

push time in an hour

push eventdanifranco/EM_Image_Segmentation

Daniel Franco

commit sha 5dbe19c9f9c2d44a4a352fcb750bc9ed22036a1f

Update Oztel code

view details

push time in an hour

issue commentijpb/MorphoLibJ

Bug in Intensity Measure 2D/3D - Max value in 32bit images

Hi, thanks for reporting! I think this may be due to a wrong initialization of the max value. I will check it!

David

ofrag

comment created time in 2 hours

startedome/omero-insight

started time in 3 hours

issue openedijpb/MorphoLibJ

Bug in Intensity Measure 2D/3D - Max value in 32bit images

Hi,

When I run Intensity Measure 2D/3D on 32 bit image that has negative values (output of FeatureJ Laplacian) with corresponding label image. The Max value is not calculated correctly for cases in which the Max is negative. a low value 4.900e-324 is given instead.

Thanks Ofra

created time in 3 hours

pull request commentfiji/jython-autocompletion

Pin to script-editor version 0.5.9

I am closing this, as this project will soon require script-editor version 0.5.10. There are two things I need from any of you (@imagejan , @frauzufall , @acardona ):

  1. Merge https://github.com/scijava/script-editor/pull/51 and
  2. Release script-editor-0.5.10

Currently, this project is pinned to a SNAPSHOT version of the editor and Travis build is failing. Once script-editor 0.5.10 is released, I will release here too, and upload both jars the the Java 8 update site . Mmmm.. Releasing inter-twined projects is indeed cumbersome.

imagejan

comment created time in 6 hours

push eventfiji/jython-autocompletion

Jan Eglinger

commit sha 4ed0695f41798fddfb03a3763e3c87358defdd66

Pin to script-editor version 0.5.9

view details

tferr

commit sha 9c3567baf01e71dafadd53f7c24651af316647cd

Replace String suggestions with CompletionText suggestions org.scijava.ui.swing.script.autocompletion.CompletionText has major advantages: It holds the replacement text plus other fields such as summary and description (populated via org.scijava.ui.swing.script.autocompletion.ClassUtil) Requires script-editor 0.5.10-SNASHOT or newer

view details

tferr

commit sha 7102a2cd9aa719e1ae3198ea30956fbcf41d8a5f

Completion list: Implement 'fuzzy search' with sorted hits This is currently only applied to methods and fields, not imports

view details

tferr

commit sha 5ce1e553341db89a46ec4b6a26a708850332b7ed

Remove unused import

view details

tferr

commit sha c74b168314d6f44594c849093b4e0186232d5fd6

Update pom

view details

tferr

commit sha 31c8f8f2f5fefd6d5a0d1c350708761b3fd2f3dd

Update parent pom to 30.0.0

view details

tferr

commit sha 5b1d5526e97dd8e257e88a52da198be066ebeb34

Update Copyright blurbs from scijava-scripts/release-version.sh bot

view details

tferr

commit sha 9fe7b5b41db11ed337e1bec2855ea6477d36a6b0

Bump script-editor dependency

view details

push time in 6 hours

pull request commentscijava/script-editor

Extend Jython auto-completion to instantiated classes

@acardona, I don't seem to have write access to this repository. Could you please merge this? Or give me write permissions? Thanks!

tferr

comment created time in 7 hours

CommitCommentEvent

push eventimagej/imagej1

Wayne Rasband

commit sha c88089f5bc37b20166756cb28af62eabeb2abf34

2021.01.24 (1.53h54; Calibration bars)

view details

push time in 10 hours

CommitCommentEvent

push eventimagej/imagej1

Wayne Rasband

commit sha f4902e5bcda49eabacb48f7e5d6933c454a9db69

2021.01.24 (1.53h53; Selection brush tool)

view details

push time in 13 hours

issue openedjantic/DeOldify

Remove Water Mark on Jupyter Lab

I successfully got DeOldify working on Jupyter Lab! Is there any way to remove the watermark from your video using that program? I could not find an option to uncheck the watermark.

created time in 16 hours

pull request commentscijava/script-editor

Extend Jython auto-completion to instantiated classes

Sounds great! On the subclasses: a one-liner filtering for dollar signs '$', indicative of subclasses, would do, at the level of the returned stream or list of autocompletions.

tferr

comment created time in 17 hours

pull request commentscijava/script-editor

Extend Jython auto-completion to instantiated classes

@acardona, sincere apologies for the delay. 'Real work' got in the way. I think I have it working, but still need to test it more seriously. Hyperlinks seem to be working, and the completion lists the type for both fields and methods. If there are no major issues I will push and release tonight.

There are still some (minor) issues like the .acess$000 entry below, but we can always tackle this at a later time point), otherwise it will never be ready image

tferr

comment created time in 19 hours

issue commentandreas128/SRFlow

Meet "NaN" Problem.

Hi @andreas128 ,

I noticed few differences in the code and paper. Can you please put some light on it.

  1. In paper actnorm and affine layers have scaling first and then the shifting is applied, whereas in the code its the other way.
  2. I also found some use of exp() switched between the calculation of value and logdets as compared to paper.

can this be possible reason for the above shown artifacts.?

Alan-xw

comment created time in 20 hours

issue closedjantic/DeOldify

Fastai Issue with DeOldify Video Colorization.

I got an error with fastai even though I installed it will conda at https://github.com/fastai/fastai. I was trying to run the video colorizer and I cannot seem to get it working. Please provide me with some info on this! Thanks! It looks like visualize.py is unable to load fastai. Screenshot 2021-01-22 224154

closed time in 2 days

TheVincinator

issue commentjantic/DeOldify

Fastai Issue with DeOldify Video Colorization.

Nevermind, I got it working!

TheVincinator

comment created time in 2 days

issue openedjantic/DeOldify

Opaque ffmpeg exception thrown running video colorization

i was just trying to run a quick test, and ffmpeg is having an issue. the error it throws is opaque and unhelpful:

Error                                     Traceback (most recent call last)

<ipython-input-30-6f2a324612f0> in <module>()
      4 
      5 if source_url is not None and source_url !='':
----> 6   video_path = colorizer.colorize_from_url(source_url, 'video.mp4', render_factor, watermarked=watermarked)
      7   show_video_in_notebook(video_path)
      8 else:

3 frames

/usr/local/lib/python3.6/dist-packages/ffmpeg/_run.py in run(stream_spec, cmd, capture_stdout, capture_stderr, input, quiet, overwrite_output)
    303     retcode = process.poll()
    304     if retcode:
--> 305         raise Error('ffmpeg', out, err)
    306     return out, err
    307 

Error: ffmpeg error (see stderr output for detail)

the unhelpful error seems to be coming from ffmpeg.

i tried to add capture_stderr=True to the run call in _extract_raw_frames, but it didn't change the output.

attempting to inspect the exception doesn't produce much more information:

source_url = 'https://youtu.be/J7GY1Xg6X20' #@param {type:"string"}
render_factor = 21  #@param {type: "slider", min: 5, max: 40}
watermarked = True #@param {type:"boolean"}
if source_url is not None and source_url !='':
  try:
    video_path = colorizer.colorize_from_url(source_url, 'video.mp4', render_factor, watermarked=watermarked)
    show_video_in_notebook(video_path)
  except Exception as e:
    print(f"Unable to colorize video: {e}")
    print(f"{e.args}")
else:
    print('Provide a video url and try again.')
Unable to colorize video: ffmpeg error (see stderr output for detail)
('ffmpeg error (see stderr output for detail)',)

any leads would be greatly appreciated!

created time in 2 days

issue commentandreas128/SRFlow

How does the Normalizing Flow loss in SRFlow behave?

Hi @andreas128,

Thanks, that is something I was looking for and it clears my doubt about the logdets term.

avinash31d

comment created time in 2 days

issue commentandreas128/SRFlow

Negative loss? is that a problem?

The final loss of SRFlow is negative and steadily decreasing:

PyTorch Normalizing Flow Loss

We did not track those two components separately. Please try to run our code and log the loss (NLL is defined here).

Did that help?

avinash31d

comment created time in 2 days

issue commentandreas128/SRFlow

x4 train

Thanks, did the pretrained models work for you?

<br>

The error occurs when back-propagating thought the 1x1 convolution.

  • Could you please post your training NLL from tensorboard?
  • If you train on a different dataset, please try to reduce the learning rate.
flybiubiu

comment created time in 2 days

issue openedandreas128/SRFlow

x4 train

Thx author!This is a great work! But I train x4 with problem. Skipping ERROR caught in nll = model.optimize_parameters(current_step): svd_cuda: the updating process of SBDSDC did not converge (error: 11)

created time in 2 days

issue commentjantic/DeOldify

Fastai Issue with DeOldify Video Colorization.

I was also wondering if it was my GPU holding me back. I am currently using the Radeon 5700 XT. Does it require an NVIDIA GPU? I found this article on it: https://www.fast.ai/2017/11/16/what-you-need/

TheVincinator

comment created time in 2 days

CommitCommentEvent

issue openedjantic/DeOldify

Fastai Issue with DeOldify Video Colorization.

I got an error with fastai even though I installed it will conda at https://github.com/fastai/fastai. I was trying to run the video colorizer and I cannot seem to get it working. Please provide me with some info on this! Thanks! It looks like visualize.py is unable to load fastai. Screenshot 2021-01-22 224154

created time in 2 days

more