1 code implementation • 12 Oct 2022 • Megh Shukla, Roshan Roy, Pankaj Singh, Shuaib Ahmed, Alexandre Alahi
We begin with a simple premise: pose estimators often predict incoherent poses for out-of-distribution samples.
1 code implementation • 27 Aug 2022 • Midhun Vayyat, Jaswin Kasi, Anuraag Bhattacharya, Shuaib Ahmed, Rahul Tallamraju
In this work, we propose CLUDA, a simple, yet novel method for performing unsupervised domain adaptation (UDA) for semantic segmentation by incorporating contrastive losses into a student-teacher learning paradigm, that makes use of pseudo-labels generated from the target domain by the teacher network.
Ranked #1 on Unsupervised Domain Adaptation on GTA5-to-Cityscapes
1 code implementation • 19 Apr 2021 • Megh Shukla, Shuaib Ahmed
We show that LearningLoss++ outperforms in identifying scenarios where the model is likely to perform poorly, which on model refinement translates into reliable performance in the open world.
no code implementations • 17 Sep 2019 • Sai Kumar Dwivedi, Vikram Gupta, Rahul Mitra, Shuaib Ahmed, Arjun Jain
To the best of our knowledge, we are the first to report the results for G-FSL and provide a strong benchmark for future research.
1 code implementation • CVPR 2019 • Devraj Mandal, Sanath Narayan, Saikumar Dwivedi, Vikram Gupta, Shuaib Ahmed, Fahad Shahbaz Khan, Ling Shao
We introduce an out-of-distribution detector that determines whether the video features belong to a seen or unseen action category.
Action Recognition In Videos Out-of-Distribution Detection +2
no code implementations • 1 Nov 2018 • Nehal Doiphode, Rahul Mitra, Shuaib Ahmed, Arjun Jain
However, just learning from covariant constraint can lead to detection of unstable features.
1 code implementation • 4 Jan 2018 • Rahul Mitra, Nehal Doiphode, Utkarsh Gautam, Sanath Narayan, Shuaib Ahmed, Sharat Chandran, Arjun Jain
Similarly on the Strecha dataset, we see an improvement of 3-5% for the matching task in non-planar scenes.
no code implementations • 24 Jan 2017 • Rahul Mitra, Jiakai Zhang, Sanath Narayan, Shuaib Ahmed, Sharat Chandran, Arjun Jain
Scenes from the Oxford ACRD, MVS and Synthetic datasets are used for evaluating the patch matching performance of the learnt descriptors while the Strecha dataset is used to evaluate the 3D reconstruction task.