Download Demo Videos
Task 1:
In this task we reconstructed color details from a foggy video. The video shows how the software works.
[download video file(.avi format, Codec: Mircrosoft Video 1 Compressor) 4.37 MB ] In this video,
the left panel shows the original video frames taken by an
industrial monitoring CCTV color camera, and the right panel shows
the see-through effect of our algorithm. The camera was pointed to
a building with windows for a while, then swayed to the left
passing a tower crane, and then swayed back. The fog is so heavy
such that some water ball became attaching to the lenses of the
CCTV camera. Observe that not only the shape, but also the colors
of objects deep into the fog were recovered.
Task 2:
In this task we reconstructed motion-sensitive details from a foggy video. The video shows how the software works.
[download video file(.avi format, Codec: Xvid ) 6.55 MB ]
In this video, the left panel and the right panel show the video source and the processing result, respectively. The red regions in the result panel mark the motion directions of objects in the dense fog.
Task 3:
In this task we reconstructed details of a city cross road from a foggy video in ITS applications such as: measure traffic flow in foggy weather, recognize vehiecle from foggy condition. The video shows how the software works.
[download video file(.avi format, Codec: Xvid ) 3.89 MB ]
In this video, the left panel and the right panel show the video source and the processing result, respectively.
Task 4:
Please note that the upper 1/3 part of this video suffers from detail loss caused by a compression algorithm in the original
video stream. If the uncompressed video stream was available, the result can improved for the upper 1/3. In this task we reconstructed details of a city cross road from a badly compressed foggy video in ITS applications such as: measure traffic flow in foggy weather, recognize vehiecle from foggy condition. The video shows how the software works.
[download video file(.avi format, Codec: Xvid ) 4.15 MB ]
In this video, the left panel and the right panel show the video source and the processing result, respectively.
For more information on this technology, please send your request to 
.