These are examples of motion segmentation using the mixture of dynamic textures. The original video is on the left, and the segmented video is on the right. The video is available in both AVI format (DivX, playable with RealPlayer or Windows Media player) and Quicktime format (H.264). A comparison with other motion segmentation methods is also available.
Segmentation of Synthetic Video
The first four synthetic examples are from Doretto, et. al., “Dynamic Texture Segmentation”, in ICCV 2003. These video are segmented using an initial contour provided in the paper.
ocean-appearance [avi (0.4 MB) | mov (0.3 MB)] Synthetic video where two patches of the ocean has been rotated 90 degrees. |
ocean-dynamics [avi (0.3 MB) | mov (0.3 MB)] Synthetic video where the dynamics of two patches of the ocean have been altered. |
ocean-fire [avi (1.2 MB) | mov (0.8 MB)] Synthetic video of two textures: ocean and fire. The segmentation follows the changing outline of the flame. |
ocean-steam [avi (1.8 MB) | mov (1.3 MB)] |
Segmentation of Synthetic Texture Database
The synthetic texture database contains 299 videos of synthetic textures with 2, 3, or 4 segments. The dataset is available here.
| synthdb2 – video textures with 2 segments. (view all) | |
texture_027 [avi (0.6 MB) | mov (0.4 MB)] |
texture_014 [avi (0.5 MB) | mov (0.4 MB)] |
texture_098 [avi (0.5 MB) | mov (0.3 MB)] |
texture_039 [avi (0.8 MB) | mov (0.6 MB)] |
| synthdb3 – video textures with 3 segments. (view all) | |
texture_003 [avi (0.8 MB) | mov (0.6 MB)] |
texture_008 [avi (0.7 MB) | mov (0.5 MB)] |
texture_031 [avi (0.3 MB) | mov (0.2 MB)] |
texture_053 [avi (0.4 MB) | mov (0.3 MB)] |
| synthdb4 – video textures with 4 segments. (view all) | |
texture_083 [avi (0.4 MB) | mov (0.2 MB)] |
texture_092 [avi (0.9 MB) | mov (0.6 MB)] |
texture_034 [avi (0.7 MB) | mov (0.5 MB)] |
texture_004 [avi (0.8 MB) | mov (0.5 MB)] |
Segmentation of Real Video
These are examples of segmenting real video using the mixture of dynamic textures.
water fountain [avi (1.0 MB) | mov (0.8 MB)] |
|
highway traffic [avi (0.2 MB) | mov (0.2 MB)] |
bridge traffic [avi (0.2 MB) | mov (0.2 MB)] |
pedestrian scene [avi (2.1 MB) | mov (1.2 MB)] |
crowded pedestrian scene [avi (3.4 MB) | mov (2.0 MB)] |
Hour-long Pedestrian Segmentation
This is a segmentation of an hour-long pedestrian video using the mixture model learned from the “crowded pedestrian scene”.
Note that this segmentation required no reinitialization at any point, or any other type of manual supervision. The sequences contain a fair variability of traffic density, various outlying events (e.g. bicyclies, skateboarders, or small vehicles, pedestrians changing direction, etc.) and variable environmental conditions (e.g. varying clouds and shadows). The video is sped up by 2 times, and each clip is about 5 minutes long. This dataset is available here.
 [avi (44 MB) | mov (24 MB)] |
 [avi (29 MB) | mov (17 MB)] |
 [avi (22 MB) | mov (13 MB)] |
 [avi (19 MB) | mov (11 MB)] |
 [avi (24 MB) | mov (14 MB)] |
 [avi (15 MB) | mov (9.5 MB)] |
Clustering Example
An example of clustering 253 traffic videos into five clusters. Six typical sequences from each cluster are shown in the five rows, which perceptually correspond to light traffic (spanning 2 clusters), medium traffic, slow traffic, and stopped traffic (“traffic jam”). This dataset is available here.
[avi (0.7 MB) | mov (0.5 MB)]
- Comparisons of motion segmentation with other methods.