Fusion4D

By Mingsong Dou, Sameh Khamis, Yu.G. Degtyarev, Philip L. Davidson, Sean Ryan Fanello, Adarsh Kowdle, Sergio Orts, Christoph Rhemann, David Kim, Jonathan Taylor and 3 more. ACM Transactions on Graphics, 2016

424 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/bf0a10f4b076cb46273c2b2ad4f3acd6f23a827e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes (2020, background): “Most successful nonrigid reconstruction systems either require RGBD data as input [7, 18, 29, 48, 80, 86], or can only reconstruct sparse geometry [56, 65, 77, 83].”
Function4D: Real-time Human Volumetric Capture from Very Sparse Consumer RGBD Sensors (2021, methodology): “To overcome this limitation, Fusion4D [10] and Motion2fusion [9] focused on real-time multi-view setups using high-end custom-design sensors, with the notion of key volume updating and learning-based surface matching.”
HyperNeRF (2021, background): “On the opposite end of the spectrum, complicated and expensive capture setups using multiple cameras and depth sensors can be used to overconstrain the problem, thereby allowing 3D scans to be registered and fused to produce high quality results [Collet et al. 2015; Dou et al. 2016].”
Co-fusion: Real-time segmentation, tracking and fusion of multiple objects (2017, methodology): “With the use of a sophisticated multi-camera rig of RGB-D sensors 4DFusion [2] can capture live deformable shapes with an exceptional level of detail and can deal with large deformations and changes in topology.”
Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans (2020, methodology): “They rely on either depth sensors [8, 14, 55] or a dense array of cameras [11, 21] to achieve the high ﬁdelity reconstruction.”
NASA: Neural Articulated Shape Approximation (2019, background): “Overall, Equation 17 allows us to design a tracking solution that directly operates on occupancy functions, without the need to compute signed distance functions [49], closest points [48], or 3D convolutions [14].”
PaMIR: Parametric Model-Conditioned Implicit Representation for Image-Based Human Reconstruction (2020, neutral): “State-of-the-art multi-view real-time systems include [22], [23].”
Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video (2020, background): “More recently, the combination of multi-view stereo with fusion algorithms integrating implicit geometry over short time windows lead to improved results and short-term temporal coherence [10, 54, 22].”
Deferred neural rendering (2019, background): “ork is that the reconstructed 3D content can be visualized by standard rendering techniques, and the approach directly generalizes to 4D capture as long as non-rigid surfaces can be reliably tracked [Dou et al. 2016; Innmann et al. 2016; Newcombe et al. 2015]. However, at the same time, imperfection”
Animatable Neural Radiance Fields for Modeling Dynamic Human Bodies (2021, methodology): “First, high-quality human reconstruction generally relies on complicated hardware, such as a dense array of cameras [56, 16] or depth sensors [10, 14].”
MaskFusion: Real-Time Recognition, Tracking and Reconstruction of Multiple Moving Objects (2018, background): “In the first case, a deformable world is assumed [12, 32, 53] and as-rigid-as-possible registration is performed, while in the second, rigid object instances are identified [40, 46] and tracked sparsely [48, 54] or densely [39].”
Immersive light field video with a layered mesh representation (2020, background): “Later efforts have extended this to provide a much larger range of motion by fusing multiple depth maps into one animated texture mesh [Collet et al. 2015; Dou et al. 2016].”
Detailed, Accurate, Human Shape Estimation from Clothed 3D Scan Sequences (2017, background): “Other work is purely model free and attempts to estimate non-rigid 3D shape over time [41, 22, 11, 25].”