• Publications
  • Influence
CNN Based Learning Using Reflection and Retinex Models for Intrinsic Image Decomposition
TLDR
In this paper, the aim is to exploit the best of the two worlds. Expand
  • 34
  • 4
  • PDF
The Second Workshop on 3D Reconstruction Meets Semantics: Challenge Results Discussion
TLDR
This paper discusses a reconstruction challenge held as a part of the second 3D Reconstruction meets Semantics workshop (3DRMS). Expand
  • 7
  • 2
  • PDF
SHREC’16 Track: Partial Shape Queries for 3D Object Retrieval
Despite numerous recent efforts, 3D object retrieval based on partial shape queries remains a challenging problem, far from being solved. The problem can be defined as: given a partial view of aExpand
  • 5
  • 1
  • PDF
Partial Shape Queries for 3D Object Retrieval
TLDR
We evaluate the performance of partial 3D object retrieval methods, for partial shape queries of various scan qualities and degrees of partiality. Expand
  • 13
Joint Learning of Intrinsic Images and Semantic Segmentation
TLDR
We propose a supervised end-to-end CNN architecture to jointly learn intrinsic image decomposition and semantic segmentation. Expand
  • 16
  • PDF
Three for one and one for three: Flow, Segmentation, and Surface Normals
TLDR
Optical flow, semantic segmentation, and surface normals represent different information modalities, yet together they bring better cues for scene understanding problems. Expand
  • 12
  • PDF
Multimodal Smart Interactive Presentation System
TLDR
The authors propose a system that allows presenters to control presentations in a natural way by their body gestures and vocal commands. Expand
  • 4
  • PDF
SIM - Smart Interactive Map with Pointing Gestures
TLDR
We propose an efficient and economic solution to transform existing non-interactive information kiosks and maps into natural interactive systems that can accept pointing gestures, a common type of gestures in communication between people. Expand
  • 3
ShadingNet: Image Intrinsics by Fine-Grained Shading Decomposition
TLDR
In general, intrinsic image decomposition algorithms interpret shading as one unified component including all photometric effects. Expand
  • 2
  • PDF
Unsupervised Generation of Optical Flow Datasets
TLDR
We present an unsupervised algorithm to generate optical flow ground truths for non-rigid movement of real-world objects, using either ground truth or predicted segmentation. Expand
  • 1
  • PDF
...
1
2
...