We present a novel real-time approach for user-guided intrinsic decomposition of static scenes captured by an RGB-D sensor. In the first step, we acquire a three-dimensional representation of the scene using a dense volumetric reconstruction framework. The obtained reconstruction serves as a proxy to densely fuse reflectance estimates and to store user-provided constraints in three-dimensional space. User constraints, in the form of constant shading and reflectance strokes, can be placed directly on the real-world geometry using an intuitive touch-based interaction metaphor, or using interactive mouse strokes. Fusing the decomposition results and constraints in three-dimensional space allows for robust propagation of this information to novel views by re-projection. We leverage this information to improve on the decomposition quality of existing intrinsic video decomposition techniques by further constraining the ill-posed decomposition problem. In addition to improved decomposition quality, we show a variety of live augmented reality applications such as recoloring of objects, relighting of scenes and editing of material appearance.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TVCG.2017.2734425 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Unfolding the hippocampus: An intrinsic coordinate system for subfield segmentations and quantitative mapping.
Neuroimage
February 2018
Brain and Mind Institute, University of Western Ontario, London, Ontario, Canada; Imaging Research Laboratories, Robarts Research Institute, University of Western Ontario, London, Ontario, Canada; Department of Medical Biophysics, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada; Department of Medical Imaging, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada.
The hippocampus, like the neocortex, has a morphological structure that is complex and variable in its folding pattern, especially in the hippocampal head. The current study presents a computational method to unfold hippocampal grey matter, with a particular focus on the hippocampal head where complexity is highest due to medial curving of the structure and the variable presence of digitations. This unfolding was performed on segmentations from high-resolution, T2-weighted 7T MRI data from 12 healthy participants and one surgical patient with epilepsy whose resected hippocampal tissue was used for histological validation.
View Article and Find Full Text PDFWe present a novel real-time approach for user-guided intrinsic decomposition of static scenes captured by an RGB-D sensor. In the first step, we acquire a three-dimensional representation of the scene using a dense volumetric reconstruction framework. The obtained reconstruction serves as a proxy to densely fuse reflectance estimates and to store user-provided constraints in three-dimensional space.
View Article and Find Full Text PDFInteractive training system for interventional electrocardiology procedures.
Med Image Anal
January 2017
Inria, France.
Recent progress in cardiac catheterization and devices has allowed the development of new therapies for severe cardiac diseases like arrhythmias and heart failure. The skills required for such interventions are very challenging to learn, and are typically acquired over several years. Virtual reality simulators may reduce this burden by allowing trainees to practice such procedures without risk to patients.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!