Search Results for author: Sujoy Paul

Found 24 papers, 9 papers with code

Analyzing the Efficacy of an LLM-Only Approach for Image-based Document Question Answering

no code implementations25 Sep 2023 Nidhi Hegde, Sujoy Paul, Gagan Madan, Gaurav Aggarwal

Recent document question answering models consist of two key components: the vision encoder, which captures layout and visual elements in images, and a Large Language Model (LLM) that helps contextualize questions to the image and supplements them with external world knowledge to generate accurate answers.

Language Modelling Large Language Model +2

Is it an i or an l: Test-time Adaptation of Text Line Recognition Models

no code implementations29 Aug 2023 Debapriya Tula, Sujoy Paul, Gagan Madan, Peter Garst, Reeve Ingle, Gaurav Aggarwal

While text line recognition models are generally trained on large corpora of real and synthetic data, such models can still make frequent mistakes if the handwriting is inscrutable or the image acquisition process adds corruptions, such as noise, blur, compression, etc.

Language Modelling Test

Novel Class Discovery without Forgetting

no code implementations21 Jul 2022 K J Joseph, Sujoy Paul, Gaurav Aggarwal, Soma Biswas, Piyush Rai, Kai Han, Vineeth N Balasubramanian

Inspired by this, we identify and formulate a new, pragmatic problem setting of NCDwF: Novel Class Discovery without Forgetting, which tasks a machine learning model to incrementally discover novel categories of instances from unlabeled data, while maintaining its performance on the previously seen categories.

Novel Class Discovery

Spacing Loss for Discovering Novel Categories

1 code implementation22 Apr 2022 K J Joseph, Sujoy Paul, Gaurav Aggarwal, Soma Biswas, Piyush Rai, Kai Han, Vineeth N Balasubramanian

Novel Class Discovery (NCD) is a learning paradigm, where a machine learning model is tasked to semantically group instances from unlabeled data, by utilizing labeled instances from a disjoint set of classes.

Novel Class Discovery

Test-time Adaptation with Slot-Centric Models

1 code implementation21 Mar 2022 Mihir Prabhudesai, Anirudh Goyal, Sujoy Paul, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gaurav Aggarwal, Thomas Kipf, Deepak Pathak, Katerina Fragkiadaki

In our work, we find evidence that these losses are insufficient for the task of scene decomposition, without also considering architectural inductive biases.

Image Classification Image Segmentation +7

SITA: Single Image Test-time Adaptation

no code implementations4 Dec 2021 Ansh Khurana, Sujoy Paul, Piyush Rai, Soma Biswas, Gaurav Aggarwal

In Test-time Adaptation (TTA), given a source model, the goal is to adapt it to make better predictions for test instances from a different distribution than the source.


Unsupervised Adaptation of Semantic Segmentation Models without Source Data

no code implementations4 Dec 2021 Sujoy Paul, Ansh Khurana, Gaurav Aggarwal

Unsupervised domain adaptation aims to adapt a model learned on the labeled source data, to a new unlabeled target dataset.

Semantic Segmentation Unsupervised Domain Adaptation

Reconstruction guided Meta-learning for Few Shot Open Set Recognition

no code implementations31 Jul 2021 Sayak Nag, Dripta S. Raychaudhuri, Sujoy Paul, Amit K. Roy-Chowdhury

However, it is a critical task in many applications like environmental monitoring, where the number of labeled examples for each class is limited.

Classification Meta-Learning +1

Cross-domain Imitation from Observations

no code implementations20 May 2021 Dripta S. Raychaudhuri, Sujoy Paul, Jeroen van Baar, Amit K. Roy-Chowdhury

Once this correspondence is found, we can directly transfer the demonstrations on one domain to the other and use it for imitation.

Imitation Learning

Unsupervised Multi-source Domain Adaptation Without Access to Source Data

1 code implementation CVPR 2021 Sk Miraj Ahmed, Dripta S. Raychaudhuri, Sujoy Paul, Samet Oymak, Amit K. Roy-Chowdhury

A recent line of work addressed this problem and proposed an algorithm that transfers knowledge to the unlabeled target domain from a single source model without requiring access to the source data.

Unsupervised Domain Adaptation

Adversarial Knowledge Transfer from Unlabeled Data

1 code implementation13 Aug 2020 Akash Gupta, Rameswar Panda, Sujoy Paul, Jianming Zhang, Amit K. Roy-Chowdhury

While machine learning approaches to visual recognition offer great promise, most of the existing methods rely heavily on the availability of large quantities of labeled training data.

Transfer Learning

Domain Adaptive Semantic Segmentation Using Weak Labels

no code implementations ECCV 2020 Sujoy Paul, Yi-Hsuan Tsai, Samuel Schulter, Amit K. Roy-Chowdhury, Manmohan Chandraker

In this work, we propose a novel framework for domain adaptation in semantic segmentation with image-level weak labels in the target domain.

Segmentation Semantic Segmentation +1

Learning from Trajectories via Subgoal Discovery

1 code implementation NeurIPS 2019 Sujoy Paul, Jeroen van Baar, Amit K. Roy-Chowdhury

Learning to solve complex goal-oriented tasks with sparse terminal-only rewards often requires an enormous number of samples.

Imitation Learning Reinforcement Learning (RL)

Context-Aware Query Selection for Active Learning in Event Recognition

no code implementations9 Apr 2019 Mahmudul Hasan, Sujoy Paul, Anastasios I. Mourikis, Amit K. Roy-Chowdhury

We formulate a conditional random field model that encodes the context and devise an information-theoretic approach that utilizes entropy and mutual information of the nodes to compute the set of most informative queries, which are labeled by a human.

Active Learning Activity Recognition +1

Weakly Supervised Video Moment Retrieval From Text Queries

1 code implementation CVPR 2019 Niluthpol Chowdhury Mithun, Sujoy Paul, Amit K. Roy-Chowdhury

The weak nature of the supervision is because, during training, we only have access to the video-text pairs rather than the temporal extent of the video to which different text descriptions relate.

Moment Retrieval Natural Language Queries +2

Trajectory-based Learning for Ball-in-Maze Games

no code implementations28 Nov 2018 Sujoy Paul, Jeroen van Baar

We show that in spite of not using human-generated trajectories and just using the simulator as a model to generate a limited number of trajectories, we can get a speed-up of about 2-3x in the learning process.

Reinforcement Learning (RL)

Incorporating Scalability in Unsupervised Spatio-Temporal Feature Learning

no code implementations6 Aug 2018 Sujoy Paul, Sourya Roy, Amit K. Roy-Chowdhury

This necessitates learning of visual features from videos in an unsupervised setting.

Adversarial Perturbations Against Real-Time Video Classification Systems

1 code implementation2 Jul 2018 Shasha Li, Ajaya Neupane, Sujoy Paul, Chengyu Song, Srikanth V. Krishnamurthy, Amit K. Roy Chowdhury, Ananthram Swami

We exploit recent advances in generative adversarial network (GAN) architectures to account for temporal correlations and generate adversarial samples that can cause misclassification rates of over 80% for targeted activities.

Classification General Classification +1

Exploiting Transitivity for Learning Person Re-Identification Models on a Budget

no code implementations CVPR 2018 Sourya Roy, Sujoy Paul, Neal E. Young, Amit K. Roy-Chowdhury

Minimization of labeling effort for person re-identification in camera networks is an important problem as most of the existing popular methods are supervised and they require large amount of manual annotations, acquiring which is a tedious job.

Person Re-Identification

The Impact of Typicality for Informative Representative Selection

no code implementations CVPR 2017 Jawadul H. Bappy, Sujoy Paul, Ertem Tuncel, Amit K. Roy-Chowdhury

In computer vision, selection of the most informative samples from a huge pool of training data in order to learn a good recognition model is an active research problem.

Active Learning Data Compression

Non-Uniform Subset Selection for Active Learning in Structured Data

1 code implementation Computer Vision and Pattern Recognition (CVPR) 2017 Sujoy Paul, Jawadul H. Bappy, Amit Roy-Chowdhury

We construct a graph from the unlabeled data to represent the underlying structure, such that each node represents a data point, and edges represent the inter-relationships between them.

Active Learning Activity Recognition +1

Cannot find the paper you are looking for? You can Submit a new open access paper.