Search Results for author: Subhransu Maji

Found 75 papers, 31 papers with code

Task2Box: Box Embeddings for Modeling Asymmetric Task Relationships

1 code implementation25 Mar 2024 Rangel Daroya, Aaron Sun, Subhransu Maji

Modeling and visualizing relationships between tasks or datasets is an important step towards solving various meta-tasks such as dataset discovery, multi-tasking, and transfer learning.

Attribute Image Classification +1

Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions

no code implementations4 Jan 2024 Oindrila Saha, Grant van Horn, Subhransu Maji

By prompting LLMs in various ways, we generate descriptions that capture visual appearance, habitat, and geographic regions and pair them with existing attributes such as the taxonomic structure of the categories.

Fine-Grained Image Classification Zero-Shot Learning

Human in-the-Loop Estimation of Cluster Count in Datasets via Similarity-Driven Nested Importance Sampling

no code implementations8 Dec 2023 Gustavo Perez, Daniel Sheldon, Grant van Horn, Subhransu Maji

Human feedback on the pairwise similarity can be used to improve the clustering, but existing approaches do not guarantee accurate count estimates.

Clustering Fine-Grained Image Classification

PARTICLE: Part Discovery and Contrastive Learning for Fine-grained Recognition

1 code implementation25 Sep 2023 Oindrila Saha, Subhransu Maji

For example, under a linear-evaluation scheme, the classification accuracy of a ResNet50 trained on ImageNet using DetCon, a self-supervised learning approach, improves from 35. 4% to 42. 0% on the Caltech-UCSD Birds, from 35. 5% to 44. 1% on the FGVC Aircraft, and from 29. 7% to 37. 4% on the Stanford Cars.

Contrastive Learning Image Classification +2

COSE: A Consistency-Sensitivity Metric for Saliency on Image Classification

1 code implementation20 Sep 2023 Rangel Daroya, Aaron Sun, Subhransu Maji

We present a set of metrics that utilize vision priors to effectively assess the performance of saliency methods on image classification tasks.

Image Classification

LU-NeRF: Scene and Pose Estimation by Synchronizing Local Unposed NeRFs

no code implementations ICCV 2023 Zezhou Cheng, Carlos Esteves, Varun Jampani, Abhishek Kar, Subhransu Maji, Ameesh Makadia

Consequently, there is growing interest in extending NeRF models to jointly optimize camera poses and scene representation, which offers an alternative to off-the-shelf SfM pipelines which have well-understood failure modes.

Pose Estimation

DISCount: Counting in Large Image Collections with Detector-Based Importance Sampling

no code implementations5 Jun 2023 Gustavo Perez, Subhransu Maji, Daniel Sheldon

Many modern applications use computer vision to detect and count objects in massive image collections.

Accidental Turntables: Learning 3D Pose by Watching Objects Turn

no code implementations13 Dec 2022 Zezhou Cheng, Matheus Gadelha, Subhransu Maji

We propose a technique for learning single-view 3D object pose estimation models by utilizing a new source of data -- in-the-wild videos where objects turn.

3D Pose Estimation

MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentation

2 code implementations18 Aug 2022 Gopal Sharma, Kangxue Yin, Subhransu Maji, Evangelos Kalogerakis, Or Litany, Sanja Fidler

As a result, the learned 2D representations are view-invariant and geometrically consistent, leading to better generalization when trained on a limited number of labeled shapes compared to alternatives that utilize self-supervision in 2D or 3D alone.

Contrastive Learning Segmentation

Cross-Modal 3D Shape Generation and Manipulation

no code implementations24 Jul 2022 Zezhou Cheng, Menglei Chai, Jian Ren, Hsin-Ying Lee, Kyle Olszewski, Zeng Huang, Subhransu Maji, Sergey Tulyakov

In this paper, we propose a generic multi-modal generative model that couples the 2D modalities and implicit 3D representations through shared latent spaces.

3D Shape Generation

Improving Few-Shot Part Segmentation using Coarse Supervision

no code implementations11 Apr 2022 Oindrila Saha, Zezhou Cheng, Subhransu Maji

A significant bottleneck in training deep networks for part segmentation is the cost of obtaining detailed annotations.

Multi-Task Learning Segmentation

How well does CLIP understand texture?

1 code implementation22 Mar 2022 Chenyun Wu, Subhransu Maji

We investigate how well CLIP understands texture in natural images described by natural language.

Material Classification Zero-Shot Learning

GANORCON: Are Generative Models Useful for Few-shot Segmentation?

no code implementations CVPR 2022 Oindrila Saha, Zezhou Cheng, Subhransu Maji

Motivated by this we present an alternative approach based on contrastive learning and compare their performance on standard few-shot part segmentation benchmarks.

Contrastive Learning Image Generation +1

Semi-Supervised Learning with Taxonomic Labels

1 code implementation23 Nov 2021 Jong-Chyi Su, Subhransu Maji

We propose techniques to incorporate coarse taxonomic labels to train image classifiers in fine-grained domains.

Transfer Learning

Domain Adaptor Networks for Hyperspectral Image Recognition

no code implementations3 Aug 2021 Gustavo Perez, Subhransu Maji

We consider the problem of adapting a network trained on three-channel color images to a hyperspectral domain with a large number of channels.

On Measuring and Controlling the Spectral Bias of the Deep Image Prior

1 code implementation2 Jul 2021 Zenglin Shi, Pascal Mettes, Subhransu Maji, 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.

Denoising Super-Resolution

The Semi-Supervised iNaturalist Challenge at the FGVC8 Workshop

2 code implementations2 Jun 2021 Jong-Chyi Su, Subhransu Maji

Semi-iNat is a challenging dataset for semi-supervised classification with a long-tailed distribution of classes, fine-grained categories, and domain shifts between labeled and unlabeled data.

A Realistic Evaluation of Semi-Supervised Learning for Fine-Grained Classification

1 code implementation CVPR 2021 Jong-Chyi Su, Zezhou Cheng, Subhransu Maji

We evaluate the effectiveness of semi-supervised learning (SSL) on a realistic benchmark where data exhibits considerable class imbalance and contains images from novel classes.

General Classification Transfer Learning

The Semi-Supervised iNaturalist-Aves Challenge at FGVC7 Workshop

2 code implementations11 Mar 2021 Jong-Chyi Su, Subhransu Maji

From this collection, we sample a subset of classes and their labels, while adding the images from the remaining classes to the unlabeled set of images.

Supervised Momentum Contrastive Learning for Few-Shot Classification

no code implementations26 Jan 2021 Orchid Majumder, Avinash Ravichandran, Subhransu Maji, Alessandro Achille, Marzia Polito, Stefano Soatto

In this work we investigate the complementary roles of these two sources of information by combining instance-discriminative contrastive learning and supervised learning in a single framework called Supervised Momentum Contrastive learning (SUPMOCO).

Classification Contrastive Learning +4

Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning

1 code implementation CVPR 2021 Zhaowei Cai, Avinash Ravichandran, Subhransu Maji, Charless Fowlkes, Zhuowen Tu, Stefano Soatto

We present a plug-in replacement for batch normalization (BN) called exponential moving average normalization (EMAN), which improves the performance of existing student-teacher based self- and semi-supervised learning techniques.

Self-Supervised Learning Semi-Supervised Image Classification

StarcNet: Machine Learning for Star Cluster Identification

1 code implementation16 Dec 2020 Gustavo Perez, Matteo Messa, Daniela Calzetti, Subhransu Maji, Dooseok Jung, Angela Adamo, Mattia Siressi

We present a machine learning (ML) pipeline to identify star clusters in the multi{color images of nearby galaxies, from observations obtained with the Hubble Space Telescope as part of the Treasury Project LEGUS (Legacy ExtraGalactic Ultraviolet Survey).

BIG-bench Machine Learning General Classification

Shot in the Dark: Few-Shot Learning with No Base-Class Labels

no code implementations6 Oct 2020 Zitian Chen, Subhransu Maji, Erik Learned-Miller

To alleviate problems caused by the distribution shift, previous research has explored the use of unlabeled examples from the novel classes, in addition to labeled examples of the base classes, which is known as the transductive setting.

Inductive Bias Self-Supervised Learning +2

PhraseCut: Language-based Image Segmentation in the Wild

1 code implementation CVPR 2020 Chenyun Wu, Zhe Lin, Scott Cohen, Trung Bui, Subhransu Maji

We consider the problem of segmenting image regions given a natural language phrase, and study it on a novel dataset of 77, 262 images and 345, 486 phrase-region pairs.

Attribute Image Segmentation +2

Describing Textures using Natural Language

no code implementations ECCV 2020 Chenyun Wu, Mikayla Timm, Subhransu Maji

Textures in natural images can be characterized by color, shape, periodicity of elements within them, and other attributes that can be described using natural language.

On Equivariant and Invariant Learning of Object Landmark Representations

1 code implementation ICCV 2021 Zezhou Cheng, Jong-Chyi Su, Subhransu Maji

Given a collection of images, humans are able to discover landmarks by modeling the shared geometric structure across instances.

Contrastive Learning Object +1

Exploring and Predicting Transferability across NLP Tasks

1 code implementation EMNLP 2020 Tu Vu, Tong Wang, Tsendsuren Munkhdalai, Alessandro Sordoni, Adam Trischler, Andrew Mattarella-Micke, Subhransu Maji, Mohit Iyyer

We also develop task embeddings that can be used to predict the most transferable source tasks for a given target task, and we validate their effectiveness in experiments controlled for source and target data size.

Language Modelling Part-Of-Speech Tagging +4

Detecting and Tracking Communal Bird Roosts in Weather Radar Data

no code implementations24 Apr 2020 Zezhou Cheng, Saadia Gabriel, Pankaj Bhambhani, Daniel Sheldon, Subhransu Maji, Andrew Laughlin, David Winkler

The US weather radar archive holds detailed information about biological phenomena in the atmosphere over the last 20 years.

Deep Manifold Prior

no code implementations8 Apr 2020 Matheus Gadelha, Rui Wang, Subhransu Maji

We present a prior for manifold structured data, such as surfaces of 3D shapes, where deep neural networks are adopted to reconstruct a target shape using gradient descent starting from a random initialization.

Denoising Gaussian Processes

Learning Generative Models of Shape Handles

no code implementations CVPR 2020 Matheus Gadelha, Giorgio Gori, Duygu Ceylan, Radomir Mech, Nathan Carr, Tamy Boubekeur, Rui Wang, Subhransu Maji

We present a generative model to synthesize 3D shapes as sets of handles -- lightweight proxies that approximate the original 3D shape -- for applications in interactive editing, shape parsing, and building compact 3D representations.

ParSeNet: A Parametric Surface Fitting Network for 3D Point Clouds

2 code implementations ECCV 2020 Gopal Sharma, Difan Liu, Subhransu Maji, Evangelos Kalogerakis, Siddhartha Chaudhuri, Radomír Měch

We propose a novel, end-to-end trainable, deep network called ParSeNet that decomposes a 3D point cloud into parametric surface patches, including B-spline patches as well as basic geometric primitives.

Neural Shape Parsers for Constructive Solid Geometry

no code implementations22 Dec 2019 Gopal Sharma, Rishabh Goyal, Difan Liu, Evangelos Kalogerakis, Subhransu Maji

We investigate two architectures for this task --- a vanilla encoder (CNN) - decoder (RNN) and another architecture that augments the encoder with an explicit memory module based on the program execution stack.

The Spectral Bias of the Deep Image Prior

1 code implementation18 Dec 2019 Prithvijit Chakrabarty, Subhransu Maji

The "deep image prior" proposed by Ulyanov et al. is an intriguing property of neural nets: a convolutional encoder-decoder network can be used as a prior for natural images.

Denoising

Learning Point Embeddings from Shape Repositories for Few-Shot Segmentation

no code implementations3 Oct 2019 Gopal Sharma, Evangelos Kalogerakis, Subhransu Maji

We present a framework for learning representations of 3D shapes that reflect the information present in this meta data and show that it leads to improved generalization for semantic segmentation tasks.

Metric Learning Segmentation +2

Visualizing and Describing Fine-grained Categories as Textures

no code implementations2 Jul 2019 Tsung-Yu Lin, Mikayla Timm, Chenyun Wu, Subhransu Maji

We analyze how categories from recent FGVC challenges can be described by their textural content.

Boosting Supervision with Self-Supervision for Few-shot Learning

no code implementations17 Jun 2019 Jong-Chyi Su, Subhransu Maji, Bharath Hariharan

We present a technique to improve the transferability of deep representations learned on small labeled datasets by introducing self-supervised tasks as auxiliary loss functions.

Few-Shot Learning Self-Supervised Learning

Inferring 3D Shapes from Image Collections using Adversarial Networks

no code implementations11 Jun 2019 Matheus Gadelha, Aartika Rai, Subhransu Maji, Rui Wang

To this end, we present new differentiable projection operators that can be used by PrGAN to learn better 3D generative models.

Generative Adversarial Network

Active Adversarial Domain Adaptation

no code implementations16 Apr 2019 Jong-Chyi Su, Yi-Hsuan Tsai, Kihyuk Sohn, Buyu Liu, Subhransu Maji, Manmohan Chandraker

Our approach, active adversarial domain adaptation (AADA), explores a duality between two related problems: adversarial domain alignment and importance sampling for adapting models across domains.

Active Learning Domain Adaptation +3

Meta-Learning with Differentiable Convex Optimization

7 code implementations CVPR 2019 Kwonjoon Lee, Subhransu Maji, Avinash Ravichandran, Stefano Soatto

We propose to use these predictors as base learners to learn representations for few-shot learning and show they offer better tradeoffs between feature size and performance across a range of few-shot recognition benchmarks.

Few-Shot Image Classification Few-Shot Learning

Task2Vec: Task Embedding for Meta-Learning

1 code implementation ICCV 2019 Alessandro Achille, Michael Lam, Rahul Tewari, Avinash Ravichandran, Subhransu Maji, Charless Fowlkes, Stefano Soatto, Pietro Perona

We demonstrate that this embedding is capable of predicting task similarities that match our intuition about semantic and taxonomic relations between different visual tasks (e. g., tasks based on classifying different types of plants are similar) We also demonstrate the practical value of this framework for the meta-task of selecting a pre-trained feature extractor for a new task.

Meta-Learning

A Deeper Look at 3D Shape Classifiers

no code implementations7 Sep 2018 Jong-Chyi Su, Matheus Gadelha, Rui Wang, Subhransu Maji

We investigate the role of representations and architectures for classifying 3D shapes in terms of their computational efficiency, generalization, and robustness to adversarial transformations.

3D Shape Classification Computational Efficiency +1

Second-order Democratic Aggregation

no code implementations ECCV 2018 Tsung-Yu Lin, Subhransu Maji, Piotr Koniusz

In this paper, we study a class of orderless aggregation functions designed to minimize interference or equalize contributions in the context of second-order features and we show that they can be computed just as efficiently as their first-order counterparts and they have favorable properties over aggregation by summation.

General Classification Material Classification +2

Multiresolution Tree Networks for 3D Point Cloud Processing

1 code implementation ECCV 2018 Matheus Gadelha, Rui Wang, Subhransu Maji

We present multiresolution tree-structured networks to process point clouds for 3D shape understanding and generation tasks.

VisemeNet: Audio-Driven Animator-Centric Speech Animation

no code implementations24 May 2018 Yang Zhou, Zhan Xu, Chris Landreth, Evangelos Kalogerakis, Subhransu Maji, Karan Singh

We present a novel deep-learning based approach to producing animator-centric speech motion curves that drive a JALI or standard FACS-based production face-rig, directly from input audio.

Graphics

SPLATNet: Sparse Lattice Networks for Point Cloud Processing

2 code implementations CVPR 2018 Hang Su, Varun Jampani, Deqing Sun, Subhransu Maji, Evangelos Kalogerakis, Ming-Hsuan Yang, Jan Kautz

We present a network architecture for processing point clouds that directly operates on a collection of points represented as a sparse set of samples in a high-dimensional lattice.

3D Part Segmentation 3D Semantic Segmentation

CSGNet: Neural Shape Parser for Constructive Solid Geometry

1 code implementation CVPR 2018 Gopal Sharma, Rishabh Goyal, Difan Liu, Evangelos Kalogerakis, Subhransu Maji

In contrast, our model uses a recurrent neural network that parses the input shape in a top-down manner, which is significantly faster and yields a compact and easy-to-interpret sequence of modeling instructions.

Reasoning about Fine-grained Attribute Phrases using Reference Games

no code implementations ICCV 2017 Jong-Chyi Su, Chenyun Wu, Huaizu Jiang, Subhransu Maji

We collect a large dataset of such phrases by asking annotators to describe several visual differences between a pair of instances within a category.

Attribute Image Retrieval +1

Improved Bilinear Pooling with CNNs

no code implementations21 Jul 2017 Tsung-Yu Lin, Subhransu Maji

We present an alternative scheme for computing gradients that is faster and yet it offers improvements over the baseline model.

Question Answering Visual Question Answering

3D Shape Reconstruction from Sketches via Multi-view Convolutional Networks

3 code implementations20 Jul 2017 Zhaoliang Lun, Matheus Gadelha, Evangelos Kalogerakis, Subhransu Maji, Rui Wang

The decoder converts this representation into depth and normal maps capturing the underlying surface from several output viewpoints.

3D Reconstruction 3D Shape Reconstruction

Shape Generation using Spatially Partitioned Point Clouds

no code implementations19 Jul 2017 Matheus Gadelha, Subhransu Maji, Rui Wang

We propose to use the expressive power of neural networks to learn a distribution over the shape coefficients in a generative-adversarial framework.

3D Shape Induction from 2D Views of Multiple Objects

no code implementations18 Dec 2016 Matheus Gadelha, Subhransu Maji, Rui Wang

In this paper we investigate the problem of inducing a distribution over three-dimensional structures given two-dimensional views of multiple objects taken from unknown viewpoints.

3D Shape Segmentation with Projective Convolutional Networks

1 code implementation CVPR 2017 Evangelos Kalogerakis, Melinos Averkiou, Subhransu Maji, Siddhartha Chaudhuri

Our architecture combines image-based Fully Convolutional Networks (FCNs) and surface-based Conditional Random Fields (CRFs) to yield coherent segmentations of 3D shapes.

Segmentation

Adapting Models to Signal Degradation using Distillation

no code implementations1 Apr 2016 Jong-Chyi Su, Subhransu Maji

Model compression and knowledge distillation have been successfully applied for cross-architecture and cross-domain transfer learning.

Domain Adaptation Knowledge Distillation +2

High Dimensional Inference with Random Maximum A-Posteriori Perturbations

no code implementations10 Feb 2016 Tamir Hazan, Francesco Orabona, Anand D. Sarwate, Subhransu Maji, Tommi Jaakkola

This paper shows that the expected value of perturb-max inference with low dimensional perturbations can be used sequentially to generate unbiased samples from the Gibbs distribution.

Vocal Bursts Intensity Prediction

Bilinear CNN Models for Fine-Grained Visual Recognition

no code implementations ICCV 2015 Tsung-Yu Lin, Aruni RoyChowdhury, Subhransu Maji

We propose bilinear models, a recognition architecture that consists of two feature extractors whose outputs are multiplied using outer product at each location of the image and pooled to obtain an image descriptor.

Fine-Grained Image Classification Fine-Grained Visual Recognition

Visualizing and Understanding Deep Texture Representations

no code implementations CVPR 2016 Tsung-Yu Lin, Subhransu Maji

A number of recent approaches have used deep convolutional neural networks (CNNs) to build texture representations.

Attribute Image Manipulation +2

Deep filter banks for texture recognition, description, and segmentation

no code implementations9 Jul 2015 Mircea Cimpoi, Subhransu Maji, Iasonas Kokkinos, 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.

Benchmarking

One-to-many face recognition with bilinear CNNs

no code implementations3 Jun 2015 Aruni RoyChowdhury, Tsung-Yu Lin, Subhransu Maji, Erik Learned-Miller

We demonstrate the performance of the B-CNN model beginning from an AlexNet-style network pre-trained on ImageNet.

Face Detection Face Model +1

Deep Filter Banks for Texture Recognition and Segmentation

1 code implementation CVPR 2015 Mircea Cimpoi, Subhransu Maji, Andrea Vedaldi

Research in texture recognition often concentrates on the problem of material recognition in uncluttered conditions, an assumption rarely met by applications.

Material Recognition Scene Recognition

Multi-view Convolutional Neural Networks for 3D Shape Recognition

no code implementations ICCV 2015 Hang Su, Subhransu Maji, Evangelos Kalogerakis, Erik Learned-Miller

A longstanding question in computer vision concerns the representation of 3D shapes for recognition: should 3D shapes be represented with descriptors operating on their native 3D formats, such as voxel grid or polygon mesh, or can they be effectively represented with view-based descriptors?

3D Point Cloud Classification 3D Shape Recognition

Bilinear CNNs for Fine-grained Visual Recognition

4 code implementations29 Apr 2015 Tsung-Yu Lin, Aruni RoyChowdhury, Subhransu Maji

We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture.

Fine-Grained Image Classification Fine-Grained Visual Recognition +1

Jointly Learning Multiple Measures of Similarities from Triplet Comparisons

no code implementations5 Mar 2015 Liwen Zhang, Subhransu Maji, Ryota Tomioka

Similarity between objects is multi-faceted and it can be easier for human annotators to measure it when the focus is on a specific aspect.

Metric Learning

Deep convolutional filter banks for texture recognition and segmentation

no code implementations25 Nov 2014 Mircea Cimpoi, Subhransu Maji, Andrea Vedaldi

Research in texture recognition often concentrates on the problem of material recognition in uncluttered conditions, an assumption rarely met by applications.

Material Recognition Scene Recognition

Understanding Objects in Detail with Fine-Grained Attributes

no code implementations CVPR 2014 Andrea Vedaldi, Siddharth Mahendran, Stavros Tsogkas, Subhransu Maji, Ross Girshick, Juho Kannala, Esa Rahtu, Iasonas Kokkinos, Matthew B. Blaschko, David Weiss, Ben Taskar, Karen Simonyan, Naomi Saphra, Sammy Mohamed

We show that the collected data can be used to study the relation between part detection and attribute prediction by diagnosing the performance of classifiers that pool information from different parts of an object.

Attribute Object +2

Similarity Comparisons for Interactive Fine-Grained Categorization

no code implementations CVPR 2014 Catherine Wah, Grant van Horn, Steve Branson, Subhransu Maji, Pietro Perona, Serge Belongie

Current human-in-the-loop fine-grained visual categorization systems depend on a predefined vocabulary of attributes and parts, usually determined by experts.

Attribute Fine-Grained Visual Categorization +3

Describing Textures in the Wild

13 code implementations CVPR 2014 Mircea Cimpoi, Subhransu Maji, Iasonas Kokkinos, Sammy Mohamed, Andrea Vedaldi

Patterns and textures are defining characteristics of many natural objects: a shirt can be striped, the wings of a butterfly can be veined, and the skin of an animal can be scaly.

Material Recognition Object Recognition

Fine-Grained Visual Classification of Aircraft

no code implementations21 Jun 2013 Subhransu Maji, Esa Rahtu, Juho Kannala, Matthew Blaschko, Andrea Vedaldi

This paper introduces FGVC-Aircraft, a new dataset containing 10, 000 images of aircraft spanning 100 aircraft models, organised in a three-level hierarchy.

Classification Fine-Grained Image Classification +1

Part Discovery from Partial Correspondence

no code implementations CVPR 2013 Subhransu Maji, Gregory Shakhnarovich

We study the problem of part discovery when partial correspondence between instances of a category are available.

object-detection Object Detection +1

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