|
|
|
|
| |
Humen Mocap Aimation Retrieval and Searching
In this paper we propose a perceptually consistent,
example-based human motion retrieval approach that is capable
of efficiently searching for and ranking similar motion sequences
given a query motion input. Our method employs a motion pattern
discovery and matching scheme that breaks human motions into
a part-based, hierarchical motion representation. Building upon
this representation, a fast string match algorithm is used for efficient
runtime motion query processing. Finally, we conducted
comparative user studies to evaluate the accuracy and perceptualconsistency
of our approach by comparing it with the state of the
art example-based human motion search algorithms.
|
|
|
| |
|
| |
Compression and Reconstruction of Motion Capture Animation |
|
| |
This paper proposes a novel scheme to compress human motion capture data based on hierarchical structure construction and motion pattern indexing. Its central scheme is to greatly reduces the redundancy in a motion sequence by utilizing the fact that the motion of a meaningful human part (e.g. arm, leg, etc.) usually exhibits similar patterns over the time. As a result, higher compression ratio has been achieved when compared with the prior art, especially for long sequences of motion capture data with repetitive motion styles. Another distinction of this work is that it provides means for flexible and intuitive global and local distortion controls.
|
|
| |
Demo (19 MB, Avi)
Paper(pdf)
Computer Graphics Forum (CGF 09) |
| |
|
| |
|
|
| |
Neighbor embedding and Example-based Super-resolution |
|
| |
|
|
| |
In a number of real-world applications, a common problem of losing sharpness occurs when we want to
increase the resolution of an digital image by simply enlarging it . Super-resolution is a solution for those problems by evaluating a high-resolution image
(HR) from one or more low-resolution images (LR). In this project we implement four neighbor embedding and example-based super-resolution solutions using some common traning image pictures. In order to obtain a HR image from given LR version, for instance, produce a 1024*768 picture from the 320*240 snapshot, the only requirement of the system is inputing some semantically similar picture pairs as training set. |
|
| |
|
| |
|
|
| |
|
