Subhransu Maji

Subhransu Maji

Auto-generated profile, based on publicly available information

Subhransu Maji

University of Massachusetts Amherst

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 January 2016 in International Journal of Computer Vision
Verified
Journal Article
Open access
PDF file
MCMircea Cimpoi
Mircea Cimpoi
SMSubhransu Maji
Subhransu Maji
IKIasonas Kokkinos
Iasonas Kokkinos
AVAndrea Vedaldi
Andrea Vedaldi
Visual textures have played a key role in image understanding because they convey important semantics of images, and because texture representations that pool local image descriptors in an orderless manner have had a tremendous impact in diverse applications. In this paper we make several contributions to texture understanding. First, instead of focusing on texture instance and material category recognition, we propose a human-interpretable vocabulary of texture attributes to describe common texture patterns, complemented by a new describable texture dataset for benchmarking. Second, we look at the problem of recognizing materials and texture attributes in realistic imaging conditions, including when textures appear in clutter, developing corresponding benchmarks on top of the recently proposed OpenSurfaces dataset. Third, we revisit classic texture represenations, including bag-of-visual-words and the Fisher vectors, in the context of deep learning and show that these have excellent efficiency and generalization properties if the convolutional layers of a deep model are used as filter banks. We obtain in this manner state-of-the-art performance in numerous datasets well beyond textures, an efficient method to apply deep features to image regions, as well as benefit in transferring features from one domain to another.

9 November 2020
Verified
Book Chapter
Closed access
PDF file
GSGopal Sharma
Gopal Sharma
DLDifan Liu
Difan Liu
SMSubhransu Maji
Subhransu Maji
EKEvangelos Kalogerakis
Evangelos Kalogerakis
SCSiddhartha Chaudhuri
Siddhartha Chaudhuri
RMRadomír Měch
Radomír Měch

16 December 2019 in Nature Climate Change
Verified
Journal Article
Closed access
PDF file
KHKyle G. Horton
Kyle G. Horton
FSFrank A. La Sorte
Frank A. La Sorte
DSDaniel Sheldon
Daniel Sheldon
TLTsung-Yu Lin
Tsung-Yu Lin
KWKevin Winner
Kevin Winner
AFAndrew Farnsworth
Andrew Farnsworth

9 November 2020
Verified
Book Chapter
Closed access
PDF file
JSJong-Chyi Su
Jong-Chyi Su
SMSubhransu Maji
Subhransu Maji
BHBharath Hariharan
Bharath Hariharan

11 February 2022 in International Journal of Computer Vision
Verified
Journal Article
Open access
PDF file
ZSZenglin Shi
Zenglin Shi
PMPascal Mettes
Pascal Mettes
SMSubhransu Maji
Subhransu Maji
CSCees G. M. Snoek
Cees G. M. Snoek
The deep image prior showed that a randomly initialized network with a suitable architecture can be trained to solve inverse imaging problems by simply optimizing it’s parameters to reconstruct a single degraded image. However, it suffers from two practical limitations. First, it remains unclear how to control the prior beyond the choice of the network architecture. Second, training requires an oracle stopping criterion as during the optimization the performance degrades after reaching an optimum value. To address these challenges we introduce a frequency-band correspondence measure to characterize the spectral bias of the deep image prior, where low-frequency image signals are learned faster and better than high-frequency counterparts. Based on our observations, we propose techniques to prevent the eventual performance degradation and accelerate convergence. We introduce a Lipschitz-controlled convolution layer and a Gaussian-controlled upsampling layer as plug-in replacements for layers used in the deep architectures. The experiments show that with these changes the performance does not degrade during optimization, relieving us from the need for an oracle stopping criterion. We further outline a stopping criterion to avoid superfluous computation. Finally, we show that our approach obtains favorable results compared to current approaches across various denoising, deblocking, inpainting, super-resolution and detail enhancement tasks. Code is available at https://github.com/shizenglin/Measure-and-Control-Spectral-Bias.

27 August 2019 in Methods in Ecology and Evolution
Verified
Journal Article
Closed access
PDF file
TLTsung‐Yu Lin
Tsung‐Yu Lin
KWKevin Winner
Kevin Winner
GBGarrett Bernstein
Garrett Bernstein
AMAbhay Mittal
Abhay Mittal
ADAdriaan M. Dokter
Adriaan M. Dokter
DSDaniel Sheldon
Daniel Sheldon