Visual-hull reconstruction from uncalibrated and unsynchronized video streams

Sinha, Sudipta N.; Pollefeys, Marc

doi:10.1109/tdpvt.2004.1335227

Cited by 18 publications

(11 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These ideas have also served as a basis for an efficient algorithm to compute an exact polyhedral visual hull , and a hybrid algorithm combining voxel carving and surface-based computation . Other recent developments include visual hull construction from uncalibrated images (Sinha and Pollefeys, 2004), as well as Bayesian (Grauman et al, 2003(Grauman et al, , 2004 and algebraic dual-space approaches to visual hull reconstruction (Brand et al, 2004). To our knowledge, our method remains the only one that computes an exact image-based combinatorial representation of the visual hull while relying solely on oriented projective constraints.…”

Section: Related Workmentioning

confidence: 99%

Projective Visual Hulls

Lazebnik

Furukawa

Ponce

2006

Int J Comput Vision

View full text Add to dashboard Cite

This article presents a novel method for computing the visual hull of a solid bounded by a smooth surface and observed by a finite set of cameras. The visual hull is the intersection of the visual cones formed by back-projecting the silhouettes found in the corresponding images. We characterize its surface as a generalized polyhedron whose faces are visual cone patches; edges are intersection curves between two viewing cones; and vertices are frontier points where the intersection of two cones is singular, or intersection points where triples of cones meet. We use the mathematical framework of oriented projective differential geometry to develop an image-based algorithm for computing the visual hull. This algorithm works in a weakly calibrated setting-that is, it only requires projective camera matrices or, equivalently, fundamental matrices for each pair of cameras. The promise of the proposed algorithm is demonstrated with experiments on several challenging data sets and a comparison to another state-of-the-art method.

show abstract

Section: Related Workmentioning

confidence: 99%

Projective Visual Hulls

Lazebnik

Furukawa

Ponce

2006

Int J Comput Vision

View full text Add to dashboard Cite

show abstract

“…Automatic synchronization methods introduce a temporal shift in the received measurements from multiple cameras during correspondence till optimal correspondence is achieved. These approaches exploit the invariance property of the projective transformations and use tracks, visual hull [35], and geometric constraints of line features [36] to rectify the temporal shift between measurements. Any remaining temporal shifts (e.g., between nonoverlapping cameras) are handled as uncertainty in the measurements during target-state estimation.…”

Section: Calibration and Synchronizationmentioning

confidence: 99%

Distributed and Decentralized Multicamera Tracking

Taj¹,

Cavallaro

2011

IEEE Signal Process. Mag.

120

View full text Add to dashboard Cite

I n recent years, the decreasing cost of cameras and advances in miniaturization have favored the deployment of largescale camera networks. This growing number of cameras enables new signal-processing applications that cooperatively use multiple sensors over wide areas. In particular, object tracking is an important step in many applications related to security, traffic monitoring, and event recognition. Such applications require the optimal tradeoff between accuracy, communication, and computing across the network. The costs associated to communication and computing depend on the type and amount of cooperation performed among cameras for information gathering, sharing, and processing to validate decisions as well as to rectify (or to reduce) estimation errors and uncertainties. In this survey, we discuss data fusion and tracking methods for camera networks and compare their performance. In particular, we cover decentralized and distributed trackers and the challenges to be addressed for the design of accurate and energy-efficient algorithms.

show abstract

“…A shape-from-silhouette (SFS) method is the most common way of converting silhouette contours into a visual hull [1,2,3,4]. Since Kanade et al proposed "Virtualized Reality" as a new visual medium for manipulating and rendering prerecorded scenes [5], many computer vision-based 3D imaging systems have been developed around the SFS approach [6,7,8].…”

Section: Introductionmentioning

confidence: 99%

“…First, most SFS methods assume perfect silhouette information of objects in advance or use simple background subtraction techniques in restricted environments such as blue screen or simple-colored backgrounds [1,2,3,7,8,10]. Chueng et al and Ishikawa et al included their own segmentation methods developed for the visual hull in a real environment to their systems [4,10], but segmentation errors directly affect the visual hull in these systems because the segmentation processes are completely independent of the reconstruction processes.…”

Section: Introductionmentioning

confidence: 99%

Compensated Visual Hull for Defective Segmentation and Occlusion

Kim¹,

Sakamoto²,

Kitahara

et al. 2007

17th International Conference on Artificial Reality and Telexistence (ICAT 2007)

View full text Add to dashboard Cite

We propose an advanced visual hull technique to compensate for outliers using the reliabilities of the silhouettes. The proposed method consists of a foreground extraction technique based on the Generalized Gaussian Family model and a compensated shape-from-silhouette algorithm. They are connected by the intra-/inter-silhouette reliabilities to compensate for carving errors from defective segmentation or partial occlusion which may occur in a real environment. The 3D reconstruction process is implemented on a graphics processing unit (GPU) to accelerate the processing speed by using the huge computational power of modern graphics hardware. Experimental results show that the proposed method provides reliable silhouette information and an accurate visual hull in real environments at a very high speed on a common PC.

show abstract

Visual-hull reconstruction from uncalibrated and unsynchronized video streams

Abstract: We present an approach for automatic reconstruction of a dynamic event using multiple video cameras recording from different viewpoints. Those

Cited by 18 publications

References 20 publications

Projective Visual Hulls

Projective Visual Hulls

Distributed and Decentralized Multicamera Tracking

Compensated Visual Hull for Defective Segmentation and Occlusion

Contact Info

Product

Resources

About