1 code implementation • 3 Apr 2024 • Harsh Rangwani, Pradipto Mondal, Mayank Mishra, Ashish Ramayee Asokan, R. Venkatesh Babu
In DeiT-LT, we introduce an efficient and effective way of distillation from CNN via distillation DIST token by using out-of-distribution images and re-weighting the distillation loss to enhance focus on tail classes.
Ranked #1 on Image Classification on iNaturalist (Overall metric)
1 code implementation • 27 Mar 2024 • Shrinivas Ramasubramanian, Harsh Rangwani, Sho Takemori, Kunal Samanta, Yuhei Umeda, Venkatesh Babu Radhakrishnan
We find that current state-of-the-art empirical techniques offer sub-optimal performance on these practical, non-decomposable performance objectives.
no code implementations • ICCV 2023 • Ankit Dhiman, Srinath R, Harsh Rangwani, Rishubh Parihar, Lokesh R Boregowda, Srinath Sridhar, R Venkatesh Babu
We propose Strata-NeRF, a single neural radiance field that implicitly captures a scene with multiple levels.
1 code implementation • 28 Apr 2023 • Harsh Rangwani, Shrinivas Ramasubramanian, Sho Takemori, Kato Takashi, Yuhei Umeda, Venkatesh Babu Radhakrishnan
Using the proposed CSST framework, we obtain practical self-training methods (for both vision and NLP tasks) for optimizing different non-decomposable metrics using deep neural networks.
1 code implementation • 20 Apr 2023 • Soumalya Nandi, Sravanti Addepalli, Harsh Rangwani, R. Venkatesh Babu
We further propose a novel \textit{training noise distribution} along with a \textit{regularized training scheme} to improve the certification within both $\ell_1$ and $\ell_2$ perturbation norms simultaneously.
1 code implementation • CVPR 2023 • Harsh Rangwani, Lavish Bansal, Kartik Sharma, Tejan Karmali, Varun Jampani, R. Venkatesh Babu
We find that one reason for degradation is the collapse of latents for each class in the $\mathcal{W}$ latent space.
Ranked #1 on Conditional Image Generation on ImageNet-LT
1 code implementation • 28 Dec 2022 • Harsh Rangwani, Sumukh K Aithal, Mayank Mishra, R. Venkatesh Babu
Real-world datasets exhibit imbalances of varying types and degrees.
Ranked #1 on Long-tail Learning on CIFAR-10-LT (ρ=50)
1 code implementation • 21 Aug 2022 • Harsh Rangwani, Naman Jaswani, Tejan Karmali, Varun Jampani, R. Venkatesh Babu
Deep long-tailed learning aims to train useful deep networks on practical, real-world imbalanced distributions, wherein most labels of the tail classes are associated with a few samples.
Ranked #1 on Image Generation on LSUN
no code implementations • 7 Aug 2022 • Tejan Karmali, Rishubh Parihar, Susmit Agrawal, Harsh Rangwani, Varun Jampani, Maneesh Singh, R. Venkatesh Babu
The quality of the generated images is predicated on two assumptions; (a) The richness of the hierarchical representations learnt by the generator, and, (b) The linearity and smoothness of the style spaces.
1 code implementation • 16 Jun 2022 • Harsh Rangwani, Sumukh K Aithal, Mayank Mishra, Arihant Jain, R. Venkatesh Babu
Based on the analysis, we introduce the Smooth Domain Adversarial Training (SDAT) procedure, which effectively enhances the performance of existing domain adversarial methods for both classification and object detection tasks.
Ranked #6 on Domain Adaptation on VisDA2017
1 code implementation • 18 Sep 2021 • Harsh Rangwani, Arihant Jain, Sumukh K Aithal, R. Venkatesh Babu
Unsupervised domain adaptation (DA) methods have focused on achieving maximal performance through aligning features from source and target domains without using labeled data in the target domain.
1 code implementation • 17 Jun 2021 • Harsh Rangwani, Konda Reddy Mopuri, R. Venkatesh Babu
However, majority of the developments focus on performance of GANs on balanced datasets.
1 code implementation • ICCV 2021 • Harsh Rangwani, Arihant Jain, Sumukh K Aithal, R. Venkatesh Babu
Unsupervised domain adaptation (DA) methods have focused on achieving maximal performance through aligning features from source and target domains without using labeled data in the target domain.
no code implementations • WS 2018 • Shashwat Trivedi, Harsh Rangwani, Anil Kumar Singh
This paper describes the best performing system for the shared task on Named Entity Recognition (NER) on code-switched data for the language pair Spanish-English (ENG-SPA).
no code implementations • SEMEVAL 2018 • Harsh Rangwani, Devang Kulshreshtha, Anil Kumar Singh
This paper describes our participation in SemEval 2018 Task 3 on Irony Detection in Tweets.