Task2Sim: Towards Effective Pre-training and Transfer from Synthetic Data We introduce a visual hallucination framework which requires only source sentences at inference time and instead uses hallucinated visual representations for multimodal machine translation. IEEE Conference on Computer Vision and Pattern Recognition ( CVPR), 2022 Yi Li, Rameswar Panda, Yoon Kim, Chun-Fu (Richard) Chen, Rogerio Feris, David Cox, Nuno Vasconcelos VALHALLA: Visual Hallucination for Machine Translation We are organizing a tutorial on Efficient Video Understanding at ICCV 2021.Two of them are selected for oral presentation. I am serving as an Area Chair for WACV 2022 and Sponsorship Chair for VCIP 2022.
We are organizing the 2nd workshop on Dynamic Neural Networks at CVPR 2022.I am also interested in image/video understanding, unsupervised/self-supervised representation learning and multimodal learning (e.g., combining vision and language). In particular, my current focus is on making AI systems more efficient, i.e., developing novel deep learning methods that can operate with less human-annotated data (data efficient), and less computation (model efficient).
My research interests mainly lie in the areas of computer vision and machine learning. Roy-Chowdhury.ĭuring Ph.D., I was very fortunate to have interned at NEC Labs, Adobe Research and Siemens Research. from UC Riverside in 2018, under the supervision of Prof.
I am a research staff member at MIT-IBM Watson AI Lab, where I work on computer vision and machine learning.