G
Gaurav Sharma
Researcher at Indian Institute of Technology Kanpur
Publications - 80
Citations - 2849
Gaurav Sharma is an academic researcher from Indian Institute of Technology Kanpur. The author has contributed to research in topics: Discriminative model & Object detection. The author has an hindex of 17, co-authored 76 publications receiving 2261 citations. Previous affiliations of Gaurav Sharma include Max Planck Society & Princeton University.
Papers
More filters
Proceedings ArticleDOI
Latent Embeddings for Zero-Shot Classification
TL;DR: This paper proposed a latent embedding model for learning a compatibility function between image and class embeddings, in the context of zero-shot classification, which augments the state-of-the-art bilinear compatibility model by incorporating latent variables.
Posted Content
Latent Embeddings for Zero-shot Classification
TL;DR: A novel latent embedding model for learning a compatibility function between image and class embeddings, in the context of zero-shot classification, that improves the state-of-the-art for various classembeddings consistently on three challenging publicly available datasets for the zero- shot setting.
Proceedings ArticleDOI
Discriminative spatial saliency for image classification
TL;DR: This work proposes to learn the discriminative spatial saliency of images while simultaneously learning a max margin classifier for a given visual classification task, and treats the saliency maps as latent variables and allow them to adapt to the image content to maximize the classification score, while regularizing the change in thesaliency maps.
Book ChapterDOI
Zero-shot object detection
TL;DR: The problem of zero-shot object detection (ZSD), which aims to detect object classes which are not observed during training, is introduced and the problems associated with selecting a background class are discussed and motivate two background-aware approaches for learning robust detectors.
Proceedings ArticleDOI
AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks for Human Action Recognition in Videos
TL;DR: The effectiveness of the proposed pooling method consistently improves on baseline pooling methods, with both RGB and optical flow based Convolutional networks, and in combination with complementary video representations is shown.