Image-based photo hulls for fast and photo-realistic new view synthesis

Slabaugh, Greg; Schafer, Ronald W.; Hans, Mat

doi:10.1016/j.rti.2003.08.004

Cited by 17 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the key ideas of our NVS method is that instead of estimating the true depth at each pixel we intend to estimate the pixel's true colour. In this respect, our method is also similar to the voxel colouring method [29], and other methods presented in [19,32]. However, one of the main differences is that our method operates mostly in the image domain, while many other methods work in the 3D object/scene domain.…”

Section: Related Workmentioning

confidence: 82%

“…Unfortunately, finding such a dense depth map is not an easy task. Many image transfer papers often assume such a dense and accurate depth map is known beforehand [32,2] or can be estimated somehow [1].…”

Section: Inverse Tensor Transfermentioning

confidence: 99%

“…The window search procedure we apply in both algorithms looks similar to that of the voxel colouring and photo-hull algorithms [29,32]. However, they assumed fully calibrated cameras and their search is mainly performed in Euclidean 3-space.…”

Section: Article In Pressmentioning

confidence: 99%

See 2 more Smart Citations

Inverse tensor transfer with applications to novel view synthesis and multi-baseline stereo

Hartley

2006

Signal Processing: Image Communication

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 82%

Section: Inverse Tensor Transfermentioning

confidence: 99%

See 1 more Smart Citation

Inverse tensor transfer with applications to novel view synthesis and multi-baseline stereo

Hartley

2006

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…Novel views of photo hulls can also be created without explicit reconstruction. Slabaugh et al [SSH03] step along each viewing ray of the target view and find a sample that Table 1 summarizes the most relevant work and compares them with our proposed new method. For more comprehensive surveys on the reconstruction of visual hulls and photo hulls, please be referred to [Dye01,SCMS01].…”

Section: Photo Hullsmentioning

confidence: 99%

Hardware‐Accelerated Rendering of Photo Hulls

Magnor

Seidel

2004

Computer Graphics Forum

View full text Add to dashboard Cite

This paper presents an efficient hardware‐accelerated method for novel view synthesis from a set of images or videos. Our method is based on the photo hull representation, which is the maximal photo‐consistent shape. We avoid the explicit reconstruction of photo hulls by adopting a view‐dependent plane‐sweeping strategy. From the target viewpoint slicing planes are rendered with reference views projected onto them. Graphics hardware is exploited to verify the photo‐consistency of each rasterized fragment. Visibilities with respect to reference views are properly modeled, and only photo‐consistent fragments are kept and colored in the target view. We present experiments with real images and animation sequences. Thanks to the more accurate shape of the photo hull representation, our method generates more realistic rendering results than methods based on visual hulls. Currently, we achieve rendering frame rates of 2–3 fps. Compared to a pure software implementation, the performance of our hardware‐accelerated method is approximately 7 times faster. Categories and Subject Descriptors (according to ACM CCS): CR Categories: I.3.3 [Computer Graphics]: Picture/Image Generation; I.3.7 [Computer Graphics]: Three‐Dimensional Graphics and Realism.

show abstract

“…Several researchers have demonstrated real-time IBR methods that can deal with dynamic scenes without off-line processing [13,9,14,[31][32][33]2,21,37]. To make the entire process on-the-fly, these methods use relatively simple geometrical models that can be constructed in real-time.…”

Section: Introductionmentioning

confidence: 99%