no code implementations • ECCV 2020 • Sayan Rakshit, Dipesh Tamboli, Pragati Shuddhodhan Meshram, Biplab Banerjee, Gemma Roig, Subhasis Chaudhuri
Besides, an adversarial learning strategy is followed to model the discriminator between the target-domain known and unknown classes.
no code implementations • 21 Nov 2024 • Sanchar Palit, Biplab Banerjee, Subhasis Chaudhuri
In the continual learning framework employing variational inference, our study introduces a regularization term that specifically targets the dynamics and population of the mean and variance of the parameters.
no code implementations • 27 Jul 2023 • Sandika Biswas, Kejie Li, Biplab Banerjee, Subhasis Chaudhuri, Hamid Rezatofighi
This paper proposes using an implicit feature representation of the scene elements to distinguish a physically plausible alignment of humans and objects from an implausible one.
no code implementations • 18 Feb 2023 • Shirsha Bose, Ritesh Sur Chowdhury, Debabrata Pal, Shivashish Bose, Biplab Banerjee, Subhasis Chaudhuri
Efficient road and building footprint extraction from satellite images are predominant in many remote sensing applications.
no code implementations • 5 Nov 2022 • Sanchar Palit, Biplab Banerjee, Subhasis Chaudhuri
We demonstrate the effectiveness of our method by showing that the embedding space remains intact after training the model with new classes and outperforms existing state-of-the-art algorithms in terms of accuracy across different sessions.
class-incremental learning
Few-Shot Class-Incremental Learning
+1
no code implementations • 28 Dec 2021 • Sayan Rakshit, Anwesh Mohanty, Ruchika Chavhan, Biplab Banerjee, Gemma Roig, Subhasis Chaudhuri
Inspired by the notion of generative feature replay, we propose a novel framework called Feature Replay based Incremental Domain Adaptation (FRIDA) which leverages a new incremental generative adversarial network (GAN) called domain-generic auxiliary classification GAN (DGAC-GAN) for producing domain-specific feature representations seamlessly.
Generative Adversarial Network
Unsupervised Domain Adaptation
no code implementations • 14 Jul 2021 • Amit More, Subhasis Chaudhuri
In this work, we propose a deep convolutional neural network (CNN) which learns to predict novel views of a scene from given collection of images.
no code implementations • 15 Feb 2021 • Priyadarshini K, Siddhartha Chaudhuri, Vivek Borkar, Subhasis Chaudhuri
To avoid redundancy between triplets, our method collectively selects batches with maximum joint entropy, which simultaneously captures both informativeness and diversity.
no code implementations • 3 Dec 2020 • Kumar Ashutosh, Saurabh Kumar, Subhasis Chaudhuri
We present a classification based approach for the next best view selection and show how we can plausibly obtain a supervisory signal for this task.
no code implementations • 17 Oct 2020 • Sayan Banerjee, S Divakar Bhat, Subhasis Chaudhuri, Rajbabu Velmurugan
Furthermore, the proposed framework does not use any semantic class labels and is entirely class agnostic.
no code implementations • 11 Oct 2020 • Ushasi Chaudhuri, Syomantak Chaudhuri, Subhasis Chaudhuri
We deal with the problem of semantic classification of challenging and highly-cluttered dataset.
no code implementations • 5 Oct 2020 • Priyadarshini K, Subhasis Chaudhuri
Discriminative features are crucial for several learning applications, such as object detection and classification.
no code implementations • 5 Oct 2020 • Ruchika Chavhan, Biplab Banerjee, Xiao Xiang Zhu, Subhasis Chaudhuri
We deal with the problem of generating textual captions from optical remote sensing (RS) images using the notion of deep reinforcement learning.
1 code implementation • 20 May 2020 • Priyadarshini K, Ritesh Goru, Siddhartha Chaudhuri, Subhasis Chaudhuri
We present an active learning strategy for training parametric models of distance metrics, given triplet-based similarity assessments: object $x_i$ is more similar to object $x_j$ than to $x_k$.
no code implementations • 9 May 2020 • Omkar Gune, Mainak Pal, Preeti Mukherjee, Biplab Banerjee, Subhasis Chaudhuri
We propose a transductive approach to reduce the effect of domain shift, where we utilize unlabeled visual data from unseen classes to generate corresponding semantic features for unseen class visual samples.
1 code implementation • 19 Feb 2020 • B. V. S Anusha, Sayan Banerjee, Subhasis Chaudhuri
These descriptors are extracted using DenseNet which significantly improves cross-sensor, cross-material and cross-dataset performance.
no code implementations • 9 Oct 2019 • T M Feroz Ali, Kalpesh K Patel, Rajbabu Velmurugan, Subhasis Chaudhuri
We derive a Mahalanobis metric induced by KFDA and argue that KFDA is efficient to be applied for metric learning in person re-identification.
no code implementations • 9 Oct 2019 • T M Feroz Ali, Subhasis Chaudhuri
In this paper, we propose a semi-supervised metric learning approach that can utilize information in unlabelled data with the help of a few labelled training samples.
3 code implementations • 25 Sep 2019 • T M Feroz Ali, Subhasis Chaudhuri
Person re-identification is the task of matching pedestrian images across non-overlapping cameras.
no code implementations • 16 Sep 2019 • Rudrajit Das, Subhasis Chaudhuri
The main result of our analysis is the derivation of a lower bound for the probability with which the inter-class distance is more than the intra-class distance in this feature space, as a function of the loss value.
no code implementations • 28 Aug 2019 • Saurabh Kumar, Biplab Banerjee, Subhasis Chaudhuri
We deal with the problem of information fusion driven satellite image/scene classification and propose a generic hallucination architecture considering that all the available sensor information are present during training while some of the image modalities may be absent while testing.
no code implementations • 12 Aug 2019 • Royston Rodrigues, Neha Bhargava, Rajbabu Velmurugan, Subhasis Chaudhuri
Typically, the methods based on this approach operate at a fixed timescale - either a single time-instant (eg.
Ranked #3 on
Anomaly Detection
on Corridor
1 code implementation • 8 May 2019 • Priyadarshini Kumari, Siddhartha Chaudhuri, Subhasis Chaudhuri
In order to design haptic icons or build a haptic vocabulary, we require a set of easily distinguishable haptic signals to avoid perceptual ambiguity, which in turn requires a way to accurately estimate the perceptual (dis)similarity of such signals.
no code implementations • ECCV 2018 • T M Feroz Ali, Subhasis Chaudhuri
In this paper we propose a novel metric learning framework called Nullspace Kernel Maximum Margin Metric Learning (NK3ML) which efficiently addresses the small sample size (SSS) problem inherent in person re-identification and offers a significant performance gain over existing state-of-the-art methods.
no code implementations • 31 Oct 2017 • Neha Bhargava, Subhasis Chaudhuri
We present an unsupervised approach to analyze crowd at various levels of granularity $-$ individual, group and collective.
no code implementations • 30 Oct 2017 • Neha Bhargava, Subhasis Chaudhuri
We show that estimation of this hidden structure corresponds to track association and group detection.