|
(An Ongoing Project…)
Image and Video Processing Lab, The
1.
Introduction
Object segmentation is
a challenge and important task in the computer vision. In the past decade, a lager
number of object segmentation methods have been proposed, meanwhile it still
lacks the reliable way for evaluating the methods’ performance. In the
traditional objective object segmentation metrics, they just use the area and
distance information to measure the difference between segmentation result
and reference (ground truth) without including any perceptual factors [1] as
shown in Fig.1. In this project, we aim at designing an objective object
segmentation metric which is derived on the basis of perceptual information
through subjective experiments and can be matched with subjective evaluation.
Fig.1 from left to right is the original
image, reference [2] and segmented object [3], respectively. The segmented
object obtains a low subjective evaluation grade, but a high grade using
traditional objective metric.
Fig.2 our subjective evaluation software 2. Group members
- Research Assistant: Ran SHI (rshi@ee.cuhk.edu.hk) - Supervisor: King Ngi Ngan (knngan@ee.cuhk.edu.hk) 3. References
[1] Elisa Drelie Gelasca
and Touradj Ebrahimi, “On
Evaluating Video Object Segmentation Quality: A Perceptually Driven Objective
Metric”, IEEE journal of selected topics in signal processing, vol. 3, no. 2,
pp. 319-335.Apr 2009. [2]
R. Achanta, S. Hemami, F.
Estrada and [3]
E. Rahtu, J. Kannala, M. Salo, and J. Heikkila,
“Segmenting salient objects from images and videos”, Proc. ECCV, pp.366-379,
Sept. 2010.
|
|
|
|
|
