Also found in: Wikipedia.
WLSDWindows Live Skydrive
WLSDWillamette Loader Steam Donkey (model trains)
References in periodicals archive ?
From Table 2, it can be seen that the bull's eye score of WLSD performs better than all the other methods, including multi-scale methods-[6][29][30].
Shape L' A ne Rouge estimates the density takes on average 2 to 3 minutes per shape, which is also much time consuming than WLSD [7].
We first test the effectiveness of the single scale WLSD by showing its bull's eye retrieval rate of each scale in Fig.
6 (a), it can be seen that the outstanding scales of WLSD are mainly focused on the low scales relatively, because the score peaks are grouped around the original point.
7 (a), the curve shows that the retrieval rate increases by iteration using WLSD in MPEG-7.
Then, we demonstrate the effectiveness of Weber's Law applied to WLSD, the experiment using only the CAF is conducted and the result is also shown in Fig.
It can be seen that the variance of the selected scale becomes larger as the number of iteration increases, and the variance along WLSD varies more dramatically than that of WLS[D.sub.w].
In the second experiment, we analysis the WLSD scale distribution.
To demonstrate the WLSD can also handle articulated shapes, we test it on relatively new Tari dataset, which consists of 1000 binary images from 50 shape categories, each category has 20 images.
From Table 3, IDSC scores higher than WLSD since it is designed to handle articulated shapes and is insensitive to articulation.
As described in the Abstract and Section 3.2.1, both the CAF and WLSD are insensitive to rotation and scale variation, we now demonstrate this property in the last experiment.
In this paper we propose a new shape descriptor based on Weber's Law, named Weber's Law Shape Descriptor (WLSD).