Publications
* indicates equal contribution.
Conference Papers
Disentangling Multiple Features in Video Sequences using Gaussian Processes in Variational Autoencoders
A variational autoencoder (MGP-VAE) which uses Gaussian processes (GP) to model the latent space for the unsupervised learning of disentangled representations in video sequences.
European Conference on Computer Vision (ECCV) 2020
Two-Step Classification using Recasted Data for Low Resource Settings
An NLI framework to improve classification performance using textual entailment.
Asia-Pacific Chapter of the Association for Computational Linguistics (AACL-IJCNLP) 2020
PrOSe: Product of Orthogonal Spheres Parameterization for Disentangled Representation Learning
Learning disentangled latent space representations as a parameterized product of orthogonal spheres for images.
British Machine Vision Conference (BMVC) 2019
DisCont: Self-Supervised Visual Attribute Disentanglement using Context Vectors
A selfsupervised framework to disentangle multiple factors of variation by exploiting the structural inductive biases within images.
PTSGM Workshop, ECCV, 2020, MLI4SD Workshop, ICML, 2020
Weakly Supervised Categoric Visual Question Generation
A weakly-supervised VAE setting for question generation using category consistent cyclic loss and structured latent space constraints.
Computer Vision and Pattern Recognition Workshop (CVPRW), 2020
Geometry of Deep Generative Models for Disentangled Representations
Analysis of latent spaces of different models of disentanglement under the lens of Riemannian Geometry.
Indian Conference on Vision, Graphics and Image Processing (ICVGIP) 2018
Learning based Methods for Code Runtime Complexity Prediction
Predicting the code complexity of programming codes using feature engineering as well as code embeddings.
European Conference on Information Retrieval (ECIR), 2020