scispace - formally typeset
T

Takeshi Naemura

Researcher at University of Tokyo

Publications -  271
Citations -  3141

Takeshi Naemura is an academic researcher from University of Tokyo. The author has contributed to research in topics: Rendering (computer graphics) & Pixel. The author has an hindex of 25, co-authored 271 publications receiving 2671 citations. Previous affiliations of Takeshi Naemura include Carnegie Mellon University & Samsung.

Papers
More filters
Proceedings ArticleDOI

Classification-Reconstruction Learning for Open-Set Recognition

TL;DR: In this article, a Classification-Reconstruction Learning for Open-Set Recognition (CROSR) method was proposed, which utilizes latent representations for reconstruction and enables robust unknown detection without harming the known class classification accuracy.
Journal ArticleDOI

3-D computer graphics based on integral photography

TL;DR: A method of synthesizing arbitrary views from IP images captured by the HDTV camera is proposed, founded on the 4-D data space Representation of light rays, which has the potential to improve the quality of images rendered by computer graphics techniques.
Journal ArticleDOI

Continuous 3D Label Stereo Matching Using Local Expansion Moves

TL;DR: This new move-making scheme is used to efficiently infer per-pixel 3D plane labels on a pairwise Markov random field (MRF) that effectively combines recently proposed slanted patch matching and curvature regularization terms.
Posted Content

Classification-Reconstruction Learning for Open-Set Recognition

TL;DR: This work utilizes latent representations for reconstruction and enables robust unknown detection without harming the known-class classification accuracy, and outperforms existing deep open-set classifiers in multiple standard datasets and is robust to diverse outliers.
Journal ArticleDOI

Real-time video-based modeling and rendering of 3D scenes

TL;DR: Using densely arranged cameras, this work has developed a system that can perform processing in real time from image pickup to interactive display, using video sequences instead of static images, at 10 frames per second.