Search Results for author: Deva Ramanan

Found 90 papers, 28 papers with code

Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories

no code implementations13 Sep 2021 Fait Poms, Vishnu Sarukkai, Ravi Teja Mullapudi, Nimit S. Sohoni, William R. Mark, Deva Ramanan, Kayvon Fatahalian

For machine learning models trained with limited labeled training data, validation stands to become the main bottleneck to reducing overall annotation costs.

FOVEA: Foveated Image Magnification for Autonomous Navigation

no code implementations27 Aug 2021 Chittesh Thavamani, Mengtian Li, Nicolas Cebron, Deva Ramanan

The detector is then fed with the warped image and we apply a differentiable backward mapping to get bounding box outputs in the original space.

Autonomous Driving Autonomous Navigation

Depth-supervised NeRF: Fewer Views and Faster Training for Free

1 code implementation6 Jul 2021 Kangle Deng, Andrew Liu, Jun-Yan Zhu, Deva Ramanan

With only two training views on real-world images, DS-NeRF significantly outperforms NeRF as well as other sparse-view variants.

Structure from Motion

Multimodal Object Detection via Bayesian Fusion

no code implementations7 Apr 2021 Yi-Ting Chen, Jinghao Shi, Christoph Mertz, Shu Kong, Deva Ramanan

Our simple approach, which we call Bayesian Fusion, is readily derived from conditional independence assumptions across different modalities.

Autonomous Vehicles Object Detection

OpenGAN: Open-Set Recognition via Open Data Generation

1 code implementation7 Apr 2021 Shu Kong, Deva Ramanan

Two conceptually elegant ideas for open-set discrimination are: 1) discriminatively learning an open-vs-closed binary discriminator by exploiting some outlier data as the open-set, and 2) unsupervised learning the closed-set data distribution with a GAN and using its discriminator as the open-set likelihood function.

Open Set Learning

Streaming Self-Training via Domain-Agnostic Unlabeled Images

no code implementations7 Apr 2021 Zhiqiu Lin, Deva Ramanan, Aayush Bansal

We present streaming self-training (SST) that aims to democratize the process of learning visual recognition models such that a non-expert user can define a new task depending on their needs via a few labeled examples and minimal domain knowledge.

Fine-Grained Image Classification Semantic Segmentation

Video Exploration via Video-Specific Autoencoders

no code implementations31 Mar 2021 Kevin Wang, Deva Ramanan, Aayush Bansal

For e. g. (1) interpolating latent codes enables temporal super-resolution and user-controllable video textures; (2) manifold reprojection enables spatial super-resolution, object removal, and denoising without training for any of the tasks.

Denoising Super-Resolution

Learning to Segment Rigid Motions from Two Frames

1 code implementation CVPR 2021 Gengshan Yang, Deva Ramanan

Geometric motion segmentation algorithms, however, generalize to novel scenes, but have yet to achieve comparable performance to appearance-based ones, due to noisy motion estimations and degenerate motion configurations.

Motion Segmentation Scene Flow Estimation

An Empirical Exploration of Open-Set Recognition via Lightweight Statistical Pipelines

no code implementations1 Jan 2021 Shu Kong, Deva Ramanan

Machine-learned safety-critical systems need to be self-aware and reliably know their unknowns in the open-world.

Open Set Learning Outlier Detection +1

Detecting Invisible People

no code implementations15 Dec 2020 Tarasha Khurana, Achal Dave, Deva Ramanan

We demonstrate that current detection and tracking systems perform dramatically worse on this task.

Monocular Depth Estimation Object Detection

Background Splitting: Finding Rare Classes in a Sea of Background

no code implementations CVPR 2021 Ravi Teja Mullapudi, Fait Poms, William R. Mark, Deva Ramanan, Kayvon Fatahalian

We focus on the real-world problem of training accurate deep models for image classification of a small number of rare categories.

Image Classification

What-If Motion Prediction for Autonomous Driving

1 code implementation24 Aug 2020 Siddhesh Khandelwal, William Qi, Jagjeet Singh, Andrew Hartnett, Deva Ramanan

Forecasting the long-term future motion of road actors is a core challenge to the deployment of safe autonomous vehicles (AVs).

Autonomous Driving motion prediction

Perceiving 3D Human-Object Spatial Arrangements from a Single Image in the Wild

1 code implementation ECCV 2020 Jason Y. Zhang, Sam Pepose, Hanbyul Joo, Deva Ramanan, Jitendra Malik, Angjoo Kanazawa

We present a method that infers spatial arrangements and shapes of humans and objects in a globally consistent 3D scene, all from a single image in-the-wild captured in an uncontrolled environment.

3D Human Pose Estimation 3D Shape Reconstruction From A Single 2D Image +2

4D Visualization of Dynamic Events from Unconstrained Multi-View Videos

no code implementations CVPR 2020 Aayush Bansal, Minh Vo, Yaser Sheikh, Deva Ramanan, Srinivasa Narasimhan

We present a data-driven approach for 4D space-time visualization of dynamic events from videos captured by hand-held multiple cameras.

Towards Streaming Perception

1 code implementation ECCV 2020 Mengtian Li, Yu-Xiong Wang, Deva Ramanan

While past work has studied the algorithmic trade-off between latency and accuracy, there has not been a clear metric to compare different methods along the Pareto optimal latency-accuracy curve.

Instance Segmentation Object Detection +1

CATER: A diagnostic dataset for Compositional Actions & TEmporal Reasoning

no code implementations ICLR 2020 Rohit Girdhar, Deva Ramanan

In this work, we build a video dataset with fully observable and controllable object and scene bias, and which truly requires spatiotemporal understanding in order to be solved.

Video Understanding

Learning Generative Models of Tissue Organization with Supervised GANs

1 code implementation31 Mar 2020 Ligong Han, Robert F. Murphy, Deva Ramanan

A key step in understanding the spatial organization of cells and tissues is the ability to construct generative models that accurately reflect that organization.

Image Generation

Unsupervised Audiovisual Synthesis via Exemplar Autoencoders

1 code implementation ICLR 2021 Kangle Deng, Aayush Bansal, Deva Ramanan

We present an unsupervised approach that converts the input speech of any individual into audiovisual streams of potentially-infinitely many output speakers.

Learning to Move with Affordance Maps

1 code implementation ICLR 2020 William Qi, Ravi Teja Mullapudi, Saurabh Gupta, Deva Ramanan

In this paper, we combine the best of both worlds with a modular approach that learns a spatial representation of a scene that is trained to be effective when coupled with traditional geometric planners.

Autonomous Navigation PointGoal Navigation

Learning to Optimally Segment Point Clouds

no code implementations10 Dec 2019 Peiyun Hu, David Held, Deva Ramanan

We prove that if we score a segmentation by the worst objectness among its individual segments, there is an efficient algorithm that finds the optimal worst-case segmentation among an exponentially large number of candidate segmentations.

Instance Segmentation Semantic Segmentation

What You See is What You Get: Exploiting Visibility for 3D Object Detection

1 code implementation CVPR 2020 Peiyun Hu, Jason Ziglar, David Held, Deva Ramanan

On the NuScenes 3D detection benchmark, we show that, by adding an additional stream for visibility input, we can significantly improve the overall detection accuracy of a state-of-the-art 3D detector.

3D Object Detection Data Augmentation

Volumetric Correspondence Networks for Optical Flow

1 code implementation NeurIPS 2019 Gengshan Yang, Deva Ramanan

As a result, SOTA networks also employ various heuristics designed to limit volumetric processing, leading to limited accuracy and overfitting.

Optical Flow Estimation

Are we asking the right questions in MovieQA?

no code implementations8 Nov 2019 Bhavan Jasani, Rohit Girdhar, Deva Ramanan

Joint vision and language tasks like visual question answering are fascinating because they explore high-level understanding, but at the same time, can be more prone to language biases.

Question Answering Visual Question Answering

Argoverse: 3D Tracking and Forecasting with Rich Maps

1 code implementation CVPR 2019 Ming-Fang Chang, John Lambert, Patsorn Sangkloy, Jagjeet Singh, Slawomir Bak, Andrew Hartnett, De Wang, Peter Carr, Simon Lucey, Deva Ramanan, James Hays

In our baseline experiments, we illustrate how detailed map information such as lane direction, driveable area, and ground height improves the accuracy of 3D object tracking and motion forecasting.

3D Object Tracking Autonomous Vehicles +3

Learning to Track Any Object

no code implementations25 Oct 2019 Achal Dave, Pavel Tokmakov, Cordelia Schmid, Deva Ramanan

Moreover, at test time the same network can be applied to detection and tracking, resulting in a unified approach for the two tasks.

Instance Segmentation Object Tracking +4

MetaPix: Few-Shot Video Retargeting

no code implementations ICLR 2020 Jessica Lee, Deva Ramanan, Rohit Girdhar

We address the task of unsupervised retargeting of human actions from one video to another.

Meta-Learning

CATER: A diagnostic dataset for Compositional Actions and TEmporal Reasoning

1 code implementation10 Oct 2019 Rohit Girdhar, Deva Ramanan

In this work, we build a video dataset with fully observable and controllable object and scene bias, and which truly requires spatiotemporal understanding in order to be solved.

Video Understanding

Growing a Brain: Fine-Tuning by Increasing Model Capacity

no code implementations CVPR 2017 Yu-Xiong Wang, Deva Ramanan, Martial Hebert

One of their remarkable properties is the ability to transfer knowledge from a large source dataset to a (typically smaller) target dataset.

Developmental Learning

Shapes and Context: In-the-Wild Image Synthesis & Manipulation

no code implementations CVPR 2019 Aayush Bansal, Yaser Sheikh, Deva Ramanan

We introduce a data-driven approach for interactively synthesizing in-the-wild images from semantic label maps.

Image Generation

Do Image Classifiers Generalize Across Time?

1 code implementation5 Jun 2019 Vaishaal Shankar, Achal Dave, Rebecca Roelofs, Deva Ramanan, Benjamin Recht, Ludwig Schmidt

Additionally, we evaluate three detection models and show that natural perturbations induce both classification as well as localization errors, leading to a median drop in detection mAP of 14 points.

General Classification Video Object Detection

Budgeted Training: Rethinking Deep Neural Network Training Under Resource Constraints

no code implementations ICLR 2020 Mengtian Li, Ersin Yumer, Deva Ramanan

We also revisit existing approaches for fast convergence and show that budget-aware learning schedules readily outperform such approaches under (the practical but under-explored) budgeted training setting.

General Classification Image Classification +5

Towards Segmenting Anything That Moves

no code implementations11 Feb 2019 Achal Dave, Pavel Tokmakov, Deva Ramanan

To address this concern, we propose two new benchmarks for generic, moving object detection, and show that our model matches top-down methods on common categories, while significantly out-performing both top-down and bottom-up methods on never-before-seen categories.

Action Detection Instance Segmentation +6

DistInit: Learning Video Representations Without a Single Labeled Video

no code implementations ICCV 2019 Rohit Girdhar, Du Tran, Lorenzo Torresani, Deva Ramanan

In this work, we propose an alternative approach to learning video representations that require no semantically labeled videos and instead leverages the years of effort in collecting and labeling large and clean still-image datasets.

Ranked #58 on Action Recognition on HMDB-51 (using extra training data)

Action Recognition Video Recognition

Photo-Sketching: Inferring Contour Drawings from Images

2 code implementations2 Jan 2019 Mengtian Li, Zhe Lin, Radomir Mech, Ersin Yumer, Deva Ramanan

Edges, boundaries and contours are important subjects of study in both computer graphics and computer vision.

Boundary Detection BSDS500

Online Model Distillation for Efficient Video Inference

1 code implementation ICCV 2019 Ravi Teja Mullapudi, Steven Chen, Keyi Zhang, Deva Ramanan, Kayvon Fatahalian

Rather than learn a specialized student model on offline data from the video stream, we train the student in an online fashion on the live video, intermittently running the teacher to provide a target for learning.

Model distillation Semantic Segmentation +2

Few-Shot Human Motion Prediction via Meta-Learning

no code implementations ECCV 2018 Liang-Yan Gui, Yu-Xiong Wang, Deva Ramanan, Jose M. F. Moura

This paper addresses the problem of few-shot human motion prediction, in the spirit of the recent progress on few-shot learning and meta-learning.

Few-Shot Learning Human motion prediction +2

Recycle-GAN: Unsupervised Video Retargeting

1 code implementation ECCV 2018 Aayush Bansal, Shugao Ma, Deva Ramanan, Yaser Sheikh

We introduce a data-driven approach for unsupervised video retargeting that translates content from one domain to another while preserving the style native to a domain, i. e., if contents of John Oliver's speech were to be transferred to Stephen Colbert, then the generated content/speech should be in Stephen Colbert's style.

Face to Face Translation Video Generation

Cross-Domain Image Matching with Deep Feature Maps

1 code implementation6 Apr 2018 Bailey Kong, James Supancic, Deva Ramanan, Charless C. Fowlkes

We investigate the problem of automatically determining what type of shoe left an impression found at a crime scene.

Image Retrieval

Active Learning with Partial Feedback

1 code implementation ICLR 2019 Peiyun Hu, Zachary C. Lipton, Anima Anandkumar, Deva Ramanan

While many active learning papers assume that the learner can simply ask for a label and receive it, real annotation often presents a mismatch between the form of a label (say, one among many classes), and the form of an annotation (typically yes/no binary feedback).

Active Learning

Brute-Force Facial Landmark Analysis With A 140,000-Way Classifier

no code implementations6 Feb 2018 Mengtian Li, Laszlo Jeni, Deva Ramanan

While most prior work treats this as a regression problem, we instead formulate it as a discrete $K$-way classification task, where a classifier is trained to return one of $K$ discrete alignments.

General Classification

Learning to Model the Tail

no code implementations NeurIPS 2017 Yu-Xiong Wang, Deva Ramanan, Martial Hebert

We cast this problem as transfer learning, where knowledge from the data-rich classes in the head of the distribution is transferred to the data-poor classes in the tail.

Few-shot Regression Image Classification +1

Patch Correspondences for Interpreting Pixel-level CNNs

no code implementations29 Nov 2017 Victor Fragoso, Chunhui Liu, Aayush Bansal, Deva Ramanan

We present compositional nearest neighbors (CompNN), a simple approach to visually interpreting distributed representations learned by a convolutional neural network (CNN) for pixel-level tasks (e. g., image synthesis and segmentation).

Image-to-Image Translation Semantic Segmentation

Attentional Pooling for Action Recognition

1 code implementation NeurIPS 2017 Rohit Girdhar, Deva Ramanan

We introduce a simple yet surprisingly powerful model to incorporate attention in action recognition and human object interaction tasks.

Action Recognition Human-Object Interaction Detection

PixelNN: Example-based Image Synthesis

no code implementations ICLR 2018 Aayush Bansal, Yaser Sheikh, Deva Ramanan

We present a simple nearest-neighbor (NN) approach that synthesizes high-frequency photorealistic images from an "incomplete" signal such as a low-resolution image, a surface normal map, or edges.

Image Generation

Learning Policies for Adaptive Tracking with Deep Feature Cascades

no code implementations ICCV 2017 Chen Huang, Simon Lucey, Deva Ramanan

Our fundamental insight is to take an adaptive approach, where easy frames are processed with cheap features (such as pixel values), while challenging frames are processed with invariant but expensive deep features.

Decision Making Visual Object Tracking

Comparing Apples and Oranges: Off-Road Pedestrian Detection on the NREC Agricultural Person-Detection Dataset

no code implementations22 Jul 2017 Zachary Pezzementi, Trenton Tabor, Peiyun Hu, Jonathan K. Chang, Deva Ramanan, Carl Wellington, Benzun P. Wisely Babu, Herman Herman

Person detection from vehicles has made rapid progress recently with the advent of multiple highquality datasets of urban and highway driving, yet no large-scale benchmark is available for the same problem in off-road or agricultural environments.

Human Detection Pedestrian Detection

Tracking as Online Decision-Making: Learning a Policy from Streaming Videos with Reinforcement Learning

no code implementations ICCV 2017 James Steven Supancic III, Deva Ramanan

We formulate tracking as an online decision-making process, where a tracking agent must follow an object despite ambiguous image frames and a limited computational budget.

Decision Making

Predictive-Corrective Networks for Action Detection

no code implementations CVPR 2017 Achal Dave, Olga Russakovsky, Deva Ramanan

While deep feature learning has revolutionized techniques for static-image understanding, the same does not quite hold for video processing.

Action Detection Optical Flow Estimation +1

ActionVLAD: Learning spatio-temporal aggregation for action classification

no code implementations CVPR 2017 Rohit Girdhar, Deva Ramanan, Abhinav Gupta, Josef Sivic, Bryan Russell

In this work, we introduce a new video representation for action classification that aggregates local convolutional features across the entire spatio-temporal extent of the video.

Action Classification Classification +2

Need for Speed: A Benchmark for Higher Frame Rate Object Tracking

1 code implementation ICCV 2017 Hamed Kiani Galoogahi, Ashton Fagg, Chen Huang, Deva Ramanan, Simon Lucey

In this paper, we propose the first higher frame rate video dataset (called Need for Speed - NfS) and benchmark for visual object tracking.

Visual Object Tracking

PixelNet: Representation of the pixels, by the pixels, and for the pixels

1 code implementation21 Feb 2017 Aayush Bansal, Xinlei Chen, Bryan Russell, Abhinav Gupta, Deva Ramanan

We explore design principles for general pixel-level prediction problems, from low-level edge detection to mid-level surface normal estimation to high-level semantic segmentation.

Edge Detection Semantic Segmentation

3D Human Pose Estimation = 2D Pose Estimation + Matching

no code implementations CVPR 2017 Ching-Hang Chen, Deva Ramanan

While many approaches try to directly predict 3D pose from image measurements, we explore a simple architecture that reasons through intermediate 2D pose predictions.

3D Human Pose Estimation 3D Pose Estimation

Tinkering Under the Hood: Interactive Zero-Shot Learning with Net Surgery

no code implementations15 Dec 2016 Vivek Krishnan, Deva Ramanan

We consider the task of visual net surgery, in which a CNN can be reconfigured without extra data to recognize novel concepts that may be omitted from the training set.

Zero-Shot Learning

Finding Tiny Faces

19 code implementations CVPR 2017 Peiyun Hu, Deva Ramanan

We explore three aspects of the problem in the context of finding small faces: the role of scale invariance, image resolution, and contextual reasoning.

Face Detection Object Recognition

PixelNet: Towards a General Pixel-level Architecture

no code implementations21 Sep 2016 Aayush Bansal, Xinlei Chen, Bryan Russell, Abhinav Gupta, Deva Ramanan

We explore architectures for general pixel-level prediction problems, from low-level edge detection to mid-level surface normal estimation to high-level semantic segmentation.

Edge Detection Semantic Segmentation

The Open World of Micro-Videos

no code implementations31 Mar 2016 Phuc Xuan Nguyen, Gregory Rogez, Charless Fowlkes, Deva Ramanan

Micro-videos are six-second videos popular on social media networks with several unique properties.

Video Understanding

Look and Think Twice: Capturing Top-Down Visual Attention With Feedback Convolutional Neural Networks

no code implementations ICCV 2015 Chunshui Cao, Xian-Ming Liu, Yi Yang, Yinan Yu, Jiang Wang, Zilei Wang, Yongzhen Huang, Liang Wang, Chang Huang, Wei Xu, Deva Ramanan, Thomas S. Huang

While feedforward deep convolutional neural networks (CNNs) have been a great success in computer vision, it is important to remember that the human visual contex contains generally more feedback connections than foward connections.

Understanding Everyday Hands in Action From RGB-D Images

no code implementations ICCV 2015 Gregory Rogez, James S. Supancic III, Deva Ramanan

We analyze functional manipulations of handheld objects, formalizing the problem as one of fine-grained grasp classification.

Bottom-Up and Top-Down Reasoning with Hierarchical Rectified Gaussians

1 code implementation CVPR 2016 Peiyun Hu, Deva Ramanan

We show that RGs can be optimized with a quadratic program (QP), that can in turn be optimized with a recurrent neural network (with rectified linear units).

First-Person Pose Recognition Using Egocentric Workspaces

no code implementations CVPR 2015 Gregory Rogez, James S. Supancic III, Deva Ramanan

In egocentric views, hands and arms are observable within a well defined volume in front of the camera.

Pose Estimation

Do We Need More Training Data?

no code implementations5 Mar 2015 Xiangxin Zhu, Carl Vondrick, Charless Fowlkes, Deva Ramanan

Datasets for training object recognition systems are steadily increasing in size.

Object Recognition

Egocentric Pose Recognition in Four Lines of Code

no code implementations29 Nov 2014 Gregory Rogez, James S. Supancic III, Deva Ramanan

We tackle the problem of estimating the 3D pose of an individual's upper limbs (arms+hands) from a chest mounted depth-camera.

Pose Estimation

Parsing Videos of Actions with Segmental Grammars

no code implementations CVPR 2014 Hamed Pirsiavash, Deva Ramanan

Real-world videos of human activities exhibit temporal structure at various scales; long videos are typically composed out of multiple action instances, where each instance is itself composed of sub-actions with variable durations and orderings.

Hierarchical structure

Parsing Occluded People

no code implementations CVPR 2014 Golnaz Ghiasi, Yi Yang, Deva Ramanan, Charless C. Fowlkes

Occlusion poses a significant difficulty for object recognition due to the combinatorial diversity of possible occlusion patterns.

Object Recognition Pose Estimation

Analysis by Synthesis: 3D Object Recognition by Object Reconstruction

no code implementations CVPR 2014 Mohsen Hejrati, Deva Ramanan

We introduce an efficient "brute-force" approach to inference that searches through a large number of candidate reconstructions, returning the optimal one.

3D Object Recognition Object Reconstruction

Capturing Long-tail Distributions of Object Subcategories

no code implementations CVPR 2014 Xiangxin Zhu, Dragomir Anguelov, Deva Ramanan

We argue that object subcategories follow a long-tail distribution: a few subcategories are common, while many are rare.

Microsoft COCO: Common Objects in Context

21 code implementations1 May 2014 Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, Piotr Dollár

We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding.

Instance Segmentation Object Localization +3

Dual coordinate solvers for large-scale structural SVMs

no code implementations6 Dec 2013 Deva Ramanan

This manuscript describes a method for training linear SVMs (including binary SVMs, SVM regression, and structural SVMs) from large, out-of-core training datasets.

3D Object Recognition Action Classification +2

Self-Paced Learning for Long-Term Tracking

no code implementations CVPR 2013 James S. Supancic III, Deva Ramanan

We address the problem of long-term object tracking, where the object may become occluded or leave-the-view.

Curriculum Learning Object Detection +1

Histograms of Sparse Codes for Object Detection

no code implementations CVPR 2013 Xiaofeng Ren, Deva Ramanan

Object detection has seen huge progress in recent years, much thanks to the heavily-engineered Histograms of Oriented Gradients (HOG) features.

Dimensionality Reduction Object Detection

Analyzing 3D Objects in Cluttered Images

no code implementations NeurIPS 2012 Mohsen Hejrati, Deva Ramanan

We use a morphable model to capture 3D within-class variation, and use a weak-perspective camera model to capture viewpoint.

3D Shape Reconstruction Viewpoint Estimation

Statistical Tests for Optimization Efficiency

no code implementations NeurIPS 2011 Levi Boyles, Anoop Korattikara, Deva Ramanan, Max Welling

Learning problems such as logistic regression are typically formulated as pure optimization problems defined on some loss function.

Bilinear classifiers for visual recognition

no code implementations NeurIPS 2009 Hamed Pirsiavash, Deva Ramanan, Charless C. Fowlkes

Bilinear classifiers are a discriminative variant of bilinear models, which capture the dependence of data on multiple factors.

Action Classification General Classification +1

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