Zobrazeno 1 - 10
of 39
pro vyhledávání: '"Kundu, Jogendra Nath"'
Autor:
Parihar, Rishubh, Bhat, Abhijnya, Basu, Abhipsa, Mallick, Saswat, Kundu, Jogendra Nath, Babu, R. Venkatesh
Diffusion Models (DMs) have emerged as powerful generative models with unprecedented image generation capability. These models are widely used for data augmentation and creative applications. However, DMs reflect the biases present in the training da
Externí odkaz:
http://arxiv.org/abs/2402.18206
Autor:
Sanyal, Sunandini, Asokan, Ashish Ramayee, Bhambri, Suvaansh, Kulkarni, Akshay, Kundu, Jogendra Nath, Babu, R. Venkatesh
Conventional Domain Adaptation (DA) methods aim to learn domain-invariant feature representations to improve the target adaptation performance. However, we motivate that domain-specificity is equally important since in-domain trained models hold cruc
Externí odkaz:
http://arxiv.org/abs/2308.14023
Autor:
Kundu, Jogendra Nath, Bhambri, Suvaansh, Kulkarni, Akshay, Sarkar, Hiran, Jampani, Varun, Babu, R. Venkatesh
Universal Domain Adaptation (UniDA) deals with the problem of knowledge transfer between two datasets with domain-shift as well as category-shift. The goal is to categorize unlabeled target samples, either into one of the "known" categories or into a
Externí odkaz:
http://arxiv.org/abs/2210.15909
Autor:
Kundu, Jogendra Nath, Bhambri, Suvaansh, Kulkarni, Akshay, Sarkar, Hiran, Jampani, Varun, Babu, R. Venkatesh
The prime challenge in unsupervised domain adaptation (DA) is to mitigate the domain shift between the source and target domains. Prior DA works show that pretext tasks could be used to mitigate this domain shift by learning domain invariant represen
Externí odkaz:
http://arxiv.org/abs/2207.13247
Autor:
Kundu, Jogendra Nath, Kulkarni, Akshay, Bhambri, Suvaansh, Mehta, Deepesh, Kulkarni, Shreyas, Jampani, Varun, Babu, R. Venkatesh
Conventional domain adaptation (DA) techniques aim to improve domain transferability by learning domain-invariant representations; while concurrently preserving the task-discriminability knowledge gathered from the labeled source data. However, the r
Externí odkaz:
http://arxiv.org/abs/2206.08009
Autor:
Kundu, Jogendra Nath, Seth, Siddharth, Jamkhandi, Anirudh, YM, Pradyumna, Jampani, Varun, Chakraborty, Anirban, Babu, R. Venkatesh
Available 3D human pose estimation approaches leverage different forms of strong (2D/3D pose) or weak (multi-view or depth) paired supervision. Barring synthetic or in-studio domains, acquiring such supervision for each new target environment is high
Externí odkaz:
http://arxiv.org/abs/2204.01971
Articulation-centric 2D/3D pose supervision forms the core training objective in most existing 3D human pose estimation techniques. Except for synthetic source environments, acquiring such rich supervision for each real target domain at deployment is
Externí odkaz:
http://arxiv.org/abs/2204.01276
Autor:
Kundu, Jogendra Nath, Seth, Siddharth, YM, Pradyumna, Jampani, Varun, Chakraborty, Anirban, Babu, R. Venkatesh
The advances in monocular 3D human pose estimation are dominated by supervised techniques that require large-scale 2D/3D pose annotations. Such methods often behave erratically in the absence of any provision to discard unfamiliar out-of-distribution
Externí odkaz:
http://arxiv.org/abs/2203.15293
Autor:
Kundu, Jogendra Nath, Kulkarni, Akshay, Bhambri, Suvaansh, Jampani, Varun, Babu, R. Venkatesh
Open compound domain adaptation (OCDA) has emerged as a practical adaptation setting which considers a single labeled source domain against a compound of multi-modal unlabeled target data in order to generalize better on novel unseen domains. We hypo
Externí odkaz:
http://arxiv.org/abs/2202.04287
Unsupervised domain adaptation (DA) has gained substantial interest in semantic segmentation. However, almost all prior arts assume concurrent access to both labeled source and unlabeled target, making them unsuitable for scenarios demanding source-f
Externí odkaz:
http://arxiv.org/abs/2108.11249