|
Silky Singh I am a Machine Learning Research Associate at MDSR Lab, Adobe. I have explored image editing, self-supervised learning, image segmentation and text-conditioned style transfer, developing a broad interest in Computer Vision and Machine Learning. In 2022, I graduated from IIT Delhi with a Bachelor's degree in Mathematics and Computing. My thesis project on "Deep Neural Network approximation for image denoising" was advised by Prof. Sivananthan Sampath. During my undergrad, I spent wonderful time at Adobe Research working with Dr. Kuldeep Kulkarni on context-aware object insertion in images, and at Max Planck Institute working on program synthesis with Prof. Eva Darulova. Email / Google Scholar / Github / LinkedIn / Misc |
|
Publications(* denotes equal contribution) |
2023 |
|
LOCATE: Self-supervised Object Discovery via Flow-guided Graph-cut and Bootstrapped Self-training
Silky Singh, Shripad Deshmukh, Mausoom Sarkar, Balaji Krishnamurthy British Machine Vision Conference (BMVC), 2023 project page / paper / arXiv / code / poster Self-supervised framework for object segmentation in both videos and images. |
|
FODVid: Flow-guided Object Discovery in Videos
Silky Singh, Shripad Deshmukh, Mausoom Sarkar, Rishabh Jain, Mayur Hemani, Balaji Krishnamurthy CVPR workshop on Learning with Limited Labelled Data for Image and Video Understanding (L3D-IVU), 2023 arXiv / paper / poster Video Object Segmentation (VOS) by leveraging flow-guided graph cut and temporal consistency in videos. |
Preprints(* denotes equal contribution) |
|
Leveraging style-based relations for text-conditioned style transfer
Surgan Jandial*, Silky Singh*, Simra Shahid*, Abhinav Java, Shripad Deshmukh |
Patents |
|
|
|
|
|
Teaching & Mentorship |
|
Last Updated: Dec 2023 |