東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

An empirical approach to grouping an...

Martin, David Royal.

Linked to FindBook

Google Book

Amazon

博客來

An empirical approach to grouping and segmentation.

Record Type:	Electronic resources : Monograph/item
Title/Author:	An empirical approach to grouping and segmentation./
Author:	Martin, David Royal.
Description:	134 p.
Notes:	Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0812.
Contained By:	Dissertation Abstracts International64-02B.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3082313

An empirical approach to grouping and segmentation.
Martin, David Royal.

An empirical approach to grouping and segmentation. - 134 p.

Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0812.

Thesis (Ph.D.)--University of California, Berkeley, 2002.

This thesis presents a novel dataset of 12,000 segmentations of 1,000 natural images by 30 human subjects. The subjects marked the locations of objects in the images, providing ground truth data for learning grouping cues and benchmarking grouping algorithms. We feel that the data-driven approach is critical for two reasons: (1) the data reflects "ecological statistics" that the human visual system has evolved to exploit, and (2) innovations in computational vision should be evaluated quantitatively.Subjects--Topical Terms:

626642
Computer Science.

An empirical approach to grouping and segmentation.
LDR:02847nmm 2200301 4500 001 1866040
005 20041220114117.5
008 130614s2002 eng d
035 $a (UnM)AAI3082313
035 $a AAI3082313
040 $a UnM $c UnM
100 1 $a Martin, David Royal. $3 1953456
245 1 3 $a An empirical approach to grouping and segmentation.
300 $a 134 p.
500 $a Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0812.
500 $a Chairs: Jitendra Malik; David Patterson.
502 $a Thesis (Ph.D.)--University of California, Berkeley, 2002.
520 $a This thesis presents a novel dataset of 12,000 segmentations of 1,000 natural images by 30 human subjects. The subjects marked the locations of objects in the images, providing ground truth data for learning grouping cues and benchmarking grouping algorithms. We feel that the data-driven approach is critical for two reasons: (1) the data reflects "ecological statistics" that the human visual system has evolved to exploit, and (2) innovations in computational vision should be evaluated quantitatively.
520 $a We develop a battery of segmentation comparison measures that we use both to validate the consistency of the human data and to provide approaches for evaluating grouping algorithms. In conjunction with the segmentation dataset, the various measures provide "micro-benchmarks" for boundary detection algorithms and pixel affinity functions, as well a benchmark for complete segmentation algorithms. Using these performance measures, we can systematically improve grouping algorithms with the human ground truth as our goal.
520 $a Starting at the lowest level, we present local boundary models based on brightness, color, and texture cues, where the cues are individually optimized with respect to the dataset and then combined in a statistically optimal manner with classifiers. The resulting detector is shown to significantly outperform prior state-of-the-art algorithms. Next, we learn from data how to combine the boundary model with patch-based features in a pixel affinity model to settle long-standing debates in computer vision with empirical results: (1) brightness boundaries are more informative than patches, and vice versa for color; (2) texture boundaries and patches are the two most powerful cues; (3) proximity is not a useful cue for grouping, it is simply a result of the process; and (4) both boundary-based and region-based approaches provide significant independent information for grouping.
590 $a School code: 0028.
650 4 $a Computer Science. $3 626642
650 4 $a Psychology, Experimental. $3 517106
690 $a 0984
690 $a 0623
710 2 0 $a University of California, Berkeley. $3 687832
773 0 $t Dissertation Abstracts International $g 64-02B.
790 1 0 $a Malik, Jitendra, $e advisor
790 1 0 $a Patterson, David, $e advisor
790 $a 0028
791 $a Ph.D.
792 $a 2002
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3082313