Uncertainty Class Activation Map (U-CAM) Using Gradient Certainty Method

Autor: Vinay P. Namboodiri, Mayank Lunayach, Badri N. Patro

Publikováno v: Patro, B, Lunayach, M & Namboodiri, V 2021, ' Uncertainty Class Activation Map (U-CAM) using Gradient Certainty method ', IEEE Transactions on Image Processing, vol. 30, pp. 1910-1924 . https://doi.org/10.1109/TIP.2020.3046916

Understanding and explaining deep learning models is an imperative task. Towards this, we propose a method that obtains gradient-based certainty estimates that also provide visual attention maps. Particularly, we solve for visual question answering t

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c6d1ed13d77a8e2215fd5099fd82dbff
https://doi.org/10.1109/tip.2020.3046916

Auto QA:The Question Is Not only What, but Also Where

Autor: Sumit Kumar, Badri N. Patro, Vinay P. Namboodiri

Publikováno v: Kumar, S, Patro, B N & Namboodiri, V P 2022, Auto QA : The Question Is Not only What, but Also Where . in 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) . vol. 2022, 9707532, Proceedings-2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022, IEEE, U. S. A., pp. 272-281, 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022, Waikoloa, USA United States, 4/01/22 . https://doi.org/10.1109/WACVW54805.2022.00033

Visual Question Answering can be a functionally relevant task if purposed as such. In this paper, we aim to investigate and evaluate its efficacy in terms of localization-based question answering. We do this specifically in the context of autonomous

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1367ac6fa562e833b2ee0392986304f8
https://purehost.bath.ac.uk/ws/files/239903566/AUTO_QA.pdf

Collaborative Learning to Generate Audio-Video Jointly

Autor: Vinay P. Namboodiri, K. S. Venkatesh, Vinod K Kurmi, Preethi Jyothi, Badri N. Patro, Vipul Bajaj

Publikováno v: ICASSP

There have been a number of techniques that have demonstrated the generation of multimedia data for one modality at a time using GANs, such as the ability to generate images, videos, and audio. However, so far, the task of multi-modal generation of d

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0c2b9f7f49d085ef6e265ecb2632926e
https://doi.org/10.1109/icassp39728.2021.9413802

Do not Forget to Attend to Uncertainty while Mitigating Catastrophic Forgetting

Autor: Vinay P. Namboodiri, Venkatesh K. Subramanian, Badri N. Patro, Vinod K Kurmi

Publikováno v: WACV

One of the major limitations of deep learning models is that they face catastrophic forgetting in an incremental learning scenario. There have been several approaches proposed to tackle the problem of incremental learning. Most of these methods are b

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::613bb2da1f84d5a83f29645e7ea25456
https://doi.org/10.1109/wacv48630.2021.00078

Visually Precise Query

Autor: Vinay P. Namboodiri, Riddhiman Dasgupta, Yokesh Kumar, Francis Tom, Sudhir Kumar, Badri N. Patro, Mithun Das Gupta

Publikováno v: ACM Multimedia

We present the problem of Visually Precise Query (VPQ) generation which enables a more intuitive match between a user's information need and an e-commerce site's product description. Given an image of a fashion item, what is the most optimum search q

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::4f9cb139760bdcdb62288d5f73612018
https://doi.org/10.1145/3394171.3413558