sleap
sleap copied to clipboard
talmolab
A deep learning framework for multi-animal pose tracking.
### Description * Add function to InferenceModel that allows saving to a frozen graph, for ingestion into a bonsai pipeline * Add standalone centroid model ### Types of changes -...
Using tracking with optical flow, similarity between frames is computed with the optical flow between the current frame and the reference frame. This is fine if the 2 frames are...
In the next release, we should aim to make our requirements for core dependencies a bit more permissive so that we don't have to uninstall stuff in Google Colab (and...
# Bug description I would like to export the labels packages (labeled + suggested or labeled + predicted + suggested would be fine). However it stops at a certain percent...
### Discussed in https://github.com/talmolab/sleap/discussions/836 Originally posted by **olinesn** July 12, 2022 Hello, Is it possible for the new frame suggestions to be added to the list, rather than to replace...
Hi, I'm attempting to train an hourglass model and am getting the error below whenever I use more than one stack. Is there a workaround? Thanks for your help! Sam...
Training fails when the project has a mix of color and grayscale videos (even when we set "Convert Image To: grayscale"). Raises the questions: - Does "Convert Image To: grayscale"...
**Problem:** We only allow for filtering based on instance score during tracking. We should have options for filtering even without tracking (e.g., during the HITL labeling process). **Solution:** Add `Predictor`...
A useful feature that would common in handy quite frequently would be a wrapper to call ffmpeg to reencode unreliably seekable videos. We currently have a help page that mentions...
**Problem**: The clip renderer supports only a subset of the visualization features in the native GUI. This is because we have to re-implement a lot of the rendering using OpenCV...
