David Russell

I think it would be useful to new users to host a completed project derived from our example videos. Then they can play around with functionality without running intensive computations...

Colorize mesh rather than ground plane based on user-calibrated image

1

New functionality allows us to select correspondences between a 3D mesh and an image and compute a camera based on this. Currently, this image is visualized projected onto the ground...

Allow selecting which camera view share intrinsics

1

As I understand it, there is currently an all-or-nothing option to couple the intrinsics of your cameras. There was interest in having a menu to select which subset of cameras...

Better validation and error messages for saving non-standard features

1

TeleSculptor is currently configured to use ORB features by default and it saves them fine. It can be configured to use other features such as SURF in the `track_features.conf` file...

This adds configs for training a pixelwise classifier based on VIAME-CSV annotations.

Make current directory evaluation in setup_viame.sh work outside of bash

1

I built VIAME but when I tried to `source setup_viame.sh` using `zsh` instead of `bash`, it failed because the `$BASH_SOURCE` variable is empty. Running `bash setup_viame.sh` runs without error, but...

[Low priority] Avoid replicating copywrite banner with outdated year

1

Low priority, but it would be nice to stop duplicating a massive header banner everywhere, or updating the year every time you touch a file. Could we have a simplified...

[FEATURE] Seek to frame

7

**Is your feature request related to a problem? If so, Please describe.** It's challenging to compare approaches when there's no way to seek by frame name or ID. More generally,...

[FEATURE] Support for semantic segmentation

3

**Is your feature request related to a problem? If so, Please describe.** I have a semantic segmentation problem pretty similar to NOAA ADAPT where I'd like to iteratively annotate data...

[FEATURE] Increase control/transparency over which annotations are visualized

4

**Is your feature request related to a problem? If so, Please describe.** It's challenging to visualize multiple annotations for the same dataset. As far as I understand, you have to...