Shlok Mishra

I'm a Research Scientist at Meta, where I work on foundational models for video and large language models (LLMs). Prior to this, I did four internships at Google where I worked in Self-Supervised Learning and Generative modelling. I received my PhD from the University of Maryland College Park where I was advised by Prof. David Jacobs.

Email / CV / Google Scholar / Twitter / Github / Linkedin

News

11/2023: Started working as Research Scientist at Meta-AI.
11/2023: Successfully defended my PhD dissertation titled "Self-Supervised Learning on large scale datasets".
02/2023: Invited for Doctoral Consortium at ICCV-2023 and BMVC 2023.
02/2023: Three papers accepted to to CVPR 2023.
02/2023: Hyperbolic Contrastive Learning is accepted to CVPR 2023.
02/2023: MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesis is accepted to CVPR 2023.
02/2023: HaLP: Hallucinating Latent Positives for Skeleton-based Self-Supervised Learning of Actions is accepted to CVPR 2023.
12/2022: Object-aware Cropping for Self-Supervised Learning is accepted to TMLR 12/2022.
10/2022: Excited to share our new paper A simple, efficient and scalable contrastive masked autoencoder for learning visual representations.
06/2022: Learning visual representations for transfer learning by suppressing texture is accepted to BMVC 2022.
06/2022: Started Internship at Google Research with Dilip Krishnan.
10/2021: Improved Presentation Attack Detection Using Image Decomposition is accepted to IJCB 2022 (ORAL).
06/2021: Started as a Research Intern at Google Research with Christian Haene.
10/2021: Robust Contrastive Learning Using Negative Samples with Diminished Semantics is accepted to Neurips 2021
10/2021: Pose and Joint-Aware Action Recognition is accepted to WACV 2022
06/2020: Started as a Research Intern at Google Research with Kuntal Sengupta.

Research

I am interested in computer vision, machine learning. Specifically, my current research interests are in Self-Supervised learning, Generative models and building models which can do learn both generative and discriminative features.

	Hyperbolic Contrastive Learning for Visual Representations beyond Objects Shlok Mishra* Songwei Ge* , Simon Kornblith, Chun-Liang Li, David Jacobs CVPR 2023 We propose a Hyperbolic loss to better encode scene and objects. arXiv code
	A simple, efficient and scalable contrastive masked autoencoder for learning visual representations Shlok Mishra, Joshua Robinson , Huiwen Chang , David Jacobs , Aaron Sarna, Aaron Maschinot, Dilip Krishnan preprint We propose a simple and scalable contrastive masked autoencoder method. arXiv code
	HaLP: Hallucinating Latent Positives for Skeleton-based Self-Supervised Learning of Actions Anshul Shah, Aniket Roy, Ketul Shah, Shlok Mishra, David Jacobs, Anoop Cherian, Rama Chellappa CVPR 2023 We hallucinate latent postives for learning skeleton encoders without labels paper
	Robust Contrastive Learning Using Negative Samples with Diminished Semantics Songwei Ge, Shlok Mishra, Chun-Liang Li , Chun-Liang Li, David Jacobs Neurips 2021* We show texture based hard negative samples can imporve generalization of contrastive learning method. paper code
	Object-Aware Cropping for Self-Supervised Learning Shlok Mishra, Anshul Shah , Ankan Bansal, Abhyuday Jagannatha, Abhishek Sharma, David Jacobs, Dilip Krishnan TMLR 12/2022 A novel cropping strategy for SSL on uncurated datasets arXiv code
	Generating Annotated Datasets via Keypoints Conditioned StyleGAN Shlok Mishra, Hossam Isack , Sergio Orts-Escolano, Luca Prasso, Rohit Pandey, Franziska Mueller, Abhimitra Meka, Jonathan Taylor, Dilip Krishnan , David Jacobs , Christian Haene. Under Submission We use StyleGAN to generate labelled data for keypoint estimation task.
	Learning Visual Representations for Transfer Learning by Suppressing Texture Shlok Mishra, Anshul Shah , Ankan Bansal, Abhyuday Jagannatha, Jonghyun Choi, Abhinav Shrivastava, Abhishek Sharma, David Jacobs BMVC 2022 Anisotropic diffusion based augmentation to reduce texture bias in supervised and self-supervised approaches paper
	Improved Presentation Attack Detection Using Image Decomposition Shlok Mishra, Kuntal Sengupta, Wen-Sheng Chu , Max Horowitz-Gelb, Sofien Bouaziz , David Jacobs IJCB 2022 (ORAL) We show albedo is a strong cue for spoof detection. arXiv
	Pose and Joint-Aware Action Recognition Anshul Shah , Shlok Mishra, Ankan Bansal, Jun-Cheng Chen, Rama Chellappa, Abhinav Shrivastava WACV 2022 We present a new model and loss for Pose-based action recognition arXiv / video / code

Service

Serving as reviewer for major machine learning and computer vision conferences (CVPR, Neurips, ICML, ICLR, AAAI, ECCV, ICCV, etc) and journals (T-PAMI, JMLR, TMLR, TIP , CVIU, IJCV). In the past I have also served in NLP conferences (ACL, EMNLP).

Template Credits : Jon Barron