A
Ajay Joshi
Researcher at Google
Publications - 22
Citations - 1152
Ajay Joshi is an academic researcher from Google. The author has contributed to research in topics: Support vector machine & Active learning. The author has an hindex of 11, co-authored 22 publications receiving 913 citations. Previous affiliations of Ajay Joshi include Mitsubishi Electric Research Laboratories & University of Minnesota.
Papers
More filters
Proceedings ArticleDOI
Multi-class active learning for image classification
TL;DR: An uncertainty measure is proposed that generalizes margin-based uncertainty to the multi-class case and is easy to compute, so that active learning can handle a large number of classes and large data sizes efficiently.
Journal ArticleDOI
Estimating pedestrian counts in groups
TL;DR: This work proposes a novel method for detecting and estimating the count of people in groups, dense or otherwise, as well as tracking them, using prior knowledge obtained from the scene and accurate camera calibration.
Journal ArticleDOI
Scalable Active Learning for Multiclass Image Classification
TL;DR: A new interaction modality for training which requires only yes-no type binary feedback instead of a precise category label is proposed, which is especially powerful in the presence of hundreds of categories.
Journal ArticleDOI
Learning to Detect Moving Shadows in Dynamic Environments
TL;DR: A novel adaptive technique for detecting moving shadows and distinguishing them from moving objects in video sequences that can dynamically adapt to varying conditions without any manual intervention, and performs better classification than previous methods on static and dynamic environments alike.
Proceedings ArticleDOI
Moving Shadow Detection with Low- and Mid-Level Reasoning
TL;DR: A multi-level shadow identification scheme which is generally applicable without restrictions on the number of light sources, illumination conditions, surface orientations, and object sizes is proposed.