izpis_h1_title_alt

Izboljšava metode za slednje objektov z dinamičnimi grafi
PETERLIN, BLAŽ (Author), Kristan, Matej (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (3,05 MB)

Abstract
Diplomska naloga obravnava izboljšavo sledilnika za kratkotrajno vizualno sledenje objektom. Sledilnik modelira sledilni objekt kot dinamični graf slikovnih regij, kjer so bližnje regije med seboj povezane. Struktura in prilagodljivost grafa sta uporabna za kvalitetno sledenje navkljub obsežnim vizualnim spremembam sledilnega objekta skozi zaporedje slik. Sledilnik se je v preteklih primerjavah s konkurenco izkazal kot kvaliteten, a z nekaj očitnimi pomanjkljivostmi, predvsem pri primerih s pogosto spreminjajočo se osvetlitvijo objekta. Na podlagi poglobljene analize sledilnika predlagamo nekaj izboljšav. Odkrili smo pomanjkljivo implementacijo v sklopu razdelitve slike na ospredje in ozadje. Predlagamo ustrezno izboljšavo, ki sledilniku poveča natančnost, predvsem v primerih hitrih sprememb osvetlitve. Poleg tega predlagamo dinamično prilagajanje omenjene razdelitve glede na velikost nastalega ospredja, ki sledilniku poveča zanesljivost trajnega sledenja tarči. Analizo in primerjavo razvitih izboljšav smo izvedli s pomočjo ocenjevalnih algoritmov iz tekmovanja VOT2015. Popravek algoritma v sklopu razdelitve slike na ospredje in ozadje izboljša natančnost sledilnika predvsem v primerih hitrih sprememb osvetlitve sledilnega objekta, hkrati pa mu izboljša zanesljivost. Obenem naredi dinamično prilagajanje razdelitve sledilnik natančnejši in precej zanesljivejši v veliki večini situacij izven hitrih sprememb osvetlitve.

Language:Slovenian
Keywords:računalniški vid, vizualno sledenje objektom, model na podlagi delcev, dinamični graf, analiza strukture graf
Work type:Undergraduate thesis (m5)
Organization:FRI - Faculty of computer and information science
Year:2016
Views:494
Downloads:339
Metadata:XML RDF-CHPDL DC-XML DC-RDF
 
Average score:(0 votes)
Your score:Voting is allowed only to logged in users.
:
Share:AddThis
AddThis uses cookies that require your consent. Edit consent...

Secondary language

Language:English
Title:An improved dynamic graph tracking algorithm
Abstract:
We propose several improvements of an existing baseline short-term visual tracking algorithm. The baseline tracker applies a dynamic graph representation to track the target. The target local parts are used as nodes in the graph, while the connections between neighboring parts represent the graph edges. This flexible model %representation of the target structure proves useful in the presence of extensive target visual changes throughout the sequence. A recent benchmark has shown that the tracker compares favorably in performance with other state-of-the-art trackers, with a notable weakness in cases of input sequences with high variance in scene and object lighting. We have performed an in-depth analysis of the tracker and propose a list of improvements. With respect to an unstable component in the tracker implementation of the foreground/background image segmentation, we propose an improvement which boosts the accuracy in cases of rapid illumination change of the target. We also propose a dynamic adjustment of the aforementioned segmentation with respect to the size of the resulting foreground, which improves tracking reliability and reduces the number of tracking failures. The implemented improvements are analyzed on the VOT2015 benchmark. Fixing the unstable component yields improvements in cases of rapid illumination change and reduces failure rate, while the dynamic segmentation adjustment improves tracking accuracy and robustness in the vast majority of cases, barring rapid illumination change.

Keywords:computer vision, visual object tracking, part-based model, dynamic graph, graph structure analysis

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Comments

Leave comment

You have to log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back