Evaluating Superpixels in Video: Metrics Beyond Figure-Ground Segmentation thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Evaluating Superpixels in Video: Metrics Beyond Figure-Ground Segmentation

Published on Apr 03, 20142879 Views

There exist almost as many superpixel segmentation algorithms as applications they can be used for. So far, the choice of the right superpixel algorithm for the task at hand is based on their abilit

Related categories

Chapter list

Evaluating Superpixels in Video: Metrics Beyond Figure-Ground Segmentation00:00
Outline00:03
What are Superpixels?00:29
What are Superpixels good for?01:34
Superpixel Algorithms01:54
Superpixel Algorithm Parameters02:13
What are desired properties of superpixel algorithms?02:25
Stability-Criteria03:56
Discontinuity-Criteria04:28
Idea: Exploit Ground Truth Optical Flow05:42
Optical Flow Datasets: KITTI06:30
Optical Flow Datasets: SINTEL07:08
So far07:42
Measuring the Segmentation Stability08:06
Measuring the Segmentation Stability: MUSE08:46
Results: Compared Algorithms09:31
Measuring the Accordance with Motion Discontinuities11:38
Measuring the Accordance with Motion Discontinuities: MDE12:01
Results: MDE12:42
Discussion of the metrics13:17
Conclusions14:06
Questions?14:52
Results: MUSE16:44