Varying delay is still a challenge to handle in the comtrol of systems with feedback delay. This paper attempt to handle varying delay by a novel approach. The control structure has been already proposed, where the sy...
详细信息
Digital photos are estimated to be generated and uploaded to social image websites (e.g. Flickr) in the amount of 375 petabytes annually [Tueld et al.], and plenty of these photos are annotated with the creation time/...
详细信息
ISBN:
(纸本)9781450319119
Digital photos are estimated to be generated and uploaded to social image websites (e.g. Flickr) in the amount of 375 petabytes annually [Tueld et al.], and plenty of these photos are annotated with the creation time/date in EXIF information. Among these photos, a large number of images are geo-tagged (e.g. GPS and landmark names) to indicate where the photo was taken. Moreover, several public databases composed of pictures captured by live cameras at various locations are available on the internet . Therefore, it has become a crucial task to enable an intuitive and efficient way for users to browse large-scale photos in space and time. Conventional photo browsing applications, e.g. ACDSee and Picasa, display photos one by one according to alphabetical or temporal order. To enrich the user's browsing experience, numerous multimedia systems, such as Photo Story, Tiling Slideshow, Auto- Collage, and IS-Scrapbook, have been developed to organize photos based on various photo metadata, including time information, visual contents, and typographical design principles. Aiming at creating a browsing style with a realistic feel, Photo Navigator is proposed to reveal the spatial relations among photos and offer a strong sense of space by taking users to fly into the scenes [Hsieh et al.]. Instead of regarding the system users as passive content receivers, Google Street View and Photo Tourism [Snavely et al.] allow the users to interact with the systems and actively determine the browsing path based on the built spatial relations among photos by the techniques of image stitching and 3D reconstruction. However, the temporal relations of photos are not well taken into account. Copyright is held by the author / owner(s).
Sports video analysis has attracted great attention in recent years. In the past decade, numerous sports video indexing approaches have been proposed at different semantic levels. In this paper, an individual level sp...
详细信息
Image re-ranking aims at improving the precision of keyword-based image retrieval, mainly by introducing visual features to re-rank. Many existing approaches require offline training for every keyword, which are unsui...
详细信息
Sports video analysis has attracted great attention in recent years. In the past decade, numerous sports video indexing approaches have been proposed at different semantic levels. In this paper, an individual level sp...
详细信息
ISBN:
(纸本)9781467316590
Sports video analysis has attracted great attention in recent years. In the past decade, numerous sports video indexing approaches have been proposed at different semantic levels. In this paper, an individual level sports video indexing (ILSVI) scheme is proposed. The individual level refers to the indexing of a sports video on a player basis, i.e. to recognize each player in a multi-player game. Since the jersey number is always "worn" by a player as the player's identity in a game, it is feasible to recognize jersey numbers for individual level indexing in sports videos. To solve the jersey number recognition problem, a principal-axis based contour descriptor is proposed. Compared to the state-of-the-art approaches, the proposed descriptor can achieve higher recognition rate and only consume much less computation power. In addition, we developed an interactive system to realize the individual level sports video indexing (ILSVI). This interactive system includes a player detection and a jersey number detection sub-systems. The interactive system can help complete the individual level sports video indexing task. We shall use basketball game videos as the basis to develop real-world systems.
Nowadays, classical 3D object management systems use only direct visible properties and common features to model relationships between objects. In this paper we propose a new Object-oriented HyperGraph-based Clusterin...
详细信息
This paper presents a novel depth map generation method based on geometric information. Our method divides a 2d image into two parts-the foreground and the background. Through extracting the predominant lines and vani...
详细信息
ISBN:
(纸本)9781467313742
This paper presents a novel depth map generation method based on geometric information. Our method divides a 2d image into two parts-the foreground and the background. Through extracting the predominant lines and vanishing point of the background, the background depth map is determined. Then we use this depth map as a `scalar' to measure the depth value of the foreground object. Finally, the depth map of the foreground and the background are integrated into one depth map by a depth fusion algorithm.
This SIG examines the vexed question of evaluation of Digital Art and how lessons on evaluation can be exchanged between the arts and mainstream HCI. We start by looking at critiques of standard approaches to evaluati...
详细信息
暂无评论