C
Cristian Sminchisescu
Researcher at Google
Publications - 189
Citations - 14699
Cristian Sminchisescu is an academic researcher from Google. The author has contributed to research in topics: Computer science & Image segmentation. The author has an hindex of 53, co-authored 173 publications receiving 12268 citations. Previous affiliations of Cristian Sminchisescu include University of Toronto & Romanian Academy.
Papers
More filters
Journal Article
Hierarchical Skeleton Abstraction
TL;DR: This work proposes a principled framework that generates a simplified, abstracted skeleton hierarchy by analyzing the quasi-stable points of a Bayesianinspired energy function, and shows that the method can produce useful multi-scale skeleton representations at a variety of abstraction levels.
Journal ArticleDOI
Discriminating Against Unrealistic Interpolations in Generative Adversarial Networks
TL;DR: It is established that the discriminator can be used effectively to avoid regions of low sample quality along shortest paths and proposed a lightweight solution for improved interpolations in pre-trained GANs.
Posted Content
Relative Flatness and Generalization
TL;DR: This paper investigated the connection between flatness and generalization by relating it to the interpolation from representative data, deriving notions of representativeness, and feature robustness, and gave rise to a natural relative flatness measure that correlates strongly with generalization, simplifies to ridge regression for ordinary least squares, and solves the reparameterization issue.
Proceedings Article
Domes to drones : Self-supervised active triangulation for 3d human pose reconstruction supplementary material
TL;DR: ACTOR as mentioned in this paper uses instance features to match 2D pose estimates in space and time using instance features and reproject 2D reprojection errors onto 2D OpenPose [2] estimates on the Panoptic test splits.
Posted Content
RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detection
Pei Sun,Weiyue Wang,Yuning Chai,Gamaleldin F. Elsayed,Alex Bewley,Xiao Zhang,Cristian Sminchisescu,Dragomir Anguelov +7 more
TL;DR: RSN as mentioned in this paper predicts foreground points from range images and applies sparse convolutions on the selected foreground points to detect objects, achieving state-of-the-art detection performance on the WOD dataset.