Abhinav Gupta - St. Stephen's College, Delhi University

Abhinav Gupta
Are you Abhinav Gupta?

Claim your profile, edit publications, add additional information:

Contact Details

Abhinav Gupta
St. Stephen's College, Delhi University
New Delhi

Pubs By Year

External Links

Pub Categories

Computer Science - Computer Vision and Pattern Recognition (42)
Computer Science - Learning (12)
Computer Science - Robotics (8)
High Energy Physics - Phenomenology (7)
Computer Science - Artificial Intelligence (3)
General Relativity and Quantum Cosmology (1)
High Energy Physics - Theory (1)
Computer Science - Multiagent Systems (1)
Computer Science - Human-Computer Interaction (1)
Computer Science - Neural and Evolutionary Computing (1)

Publications Authored By Abhinav Gupta

A crucial capability of real-world intelligent agents is their ability to plan a sequence of actions to achieve their goals in the visual world. In this work, we address the problem of visual semantic planning: the task of predicting a sequence of actions from visual observations that transform a dynamic environment from an initial state to a goal state. Doing so entails knowledge about objects and their affordances, as well as actions and their preconditions and effects. Read More

We present the 2017 WebVision Challenge, a public image recognition challenge designed for deep learning based on web images without instance-level human annotation. Following the spirit of previous vision challenges, such as ILSVRC, Places2 and PASCAL VOC, which have played critical roles in the development of computer vision by contributing to the community with large scale annotated data for model designing and standardized benchmarking, we contribute with this challenge a large scale web images dataset, and a public competition with a workshop co-located with CVPR 2017. The WebVision dataset contains more than $2. Read More

Current approaches in video forecasting attempt to generate videos directly in pixel space using Generative Adversarial Networks (GANs) or Variational Autoencoders (VAEs). However, since these approaches try to model all the structure and scene dynamics at once, in unconstrained settings they often generate uninterpretable results. Our insight is to model the forecasting problem at a higher level of abstraction. Read More

How do you learn to navigate an Unmanned Aerial Vehicle (UAV) and avoid obstacles? One approach is to use a small dataset collected by human experts: however, high capacity learning algorithms tend to overfit when trained with little data. An alternative is to use simulation. But the gap between simulation and real world remains large especially for perception problems. Read More

Modeling instance-level context and object-object relationships is extremely challenging. It requires reasoning about bounding boxes of different classes, locations \etc. Above all, instance-level spatial reasoning inherently requires modeling conditional distributions on previous detections. Read More

Collecting fully annotated image datasets is challenging and expensive. Many types of weak supervision have been explored: weak manual annotations, web search results, temporal continuity, ambient sound and others. We focus on one particular unexplored mode: visual questions that are asked about images. Read More

How do we learn an object detector that is invariant to occlusions and deformations? Our current solution is to use a data-driven strategy -- collect large-scale datasets which have object instances under different conditions. The hope is that the final classifier can use these examples to learn invariances. But is it really possible to see all the occlusions in a dataset? We argue that like categories, occlusions and object deformations also follow a long-tail. Read More

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. We do so by integrating state-of-the-art two-stream networks with learnable spatio-temporal feature aggregation. The resulting architecture is end-to-end trainable for whole-video classification. Read More

Deep neural networks coupled with fast simulation and improved computation have led to recent successes in the field of reinforcement learning (RL). However, most current RL-based approaches fail to generalize since: (a) the gap between simulation and real world is so large that policy-learning approaches fail to transfer; (b) even if policy learning is done in real world, the data scarcity leads to failed generalization from training to test scenarios (e.g. Read More

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. Convolutional predictors, such as the fully-convolutional network (FCN), have achieved remarkable success by exploiting the spatial redundancy of neighboring pixels through convolutional processing. Though computationally efficient, we point out that such approaches are not statistically efficient during learning precisely because spatial redundancy limits the information learned from neighboring pixels. Read More

We adapted the join-training scheme of Faster RCNN framework from Caffe to TensorFlow as a baseline implementation for object detection. Our code is made publicly available. This report documents the simplifications made to the original pipeline, with justifications from ablation analysis on both PASCAL VOC 2007 and COCO 2014. Read More

We present an approach to effectively use millions of images with noisy annotations in conjunction with a small subset of cleanly-annotated images to learn powerful image representations. One common approach to combine clean and noisy data is to first pre-train a network using the large noisy dataset and then fine-tune with the clean dataset. We show this approach does not fully leverage the information contained in the clean set. Read More

In this paper we investigate 3D shape attributes as a means to understand the shape of an object in a single image. To this end, we make a number of contributions: (i) we introduce and define a set of 3D shape attributes, including planarity, symmetry and occupied space; (ii) we show that such properties can be successfully inferred from a single image using a Convolutional Neural Network (CNN); (iii) we introduce a 143K image dataset of sculptures with 2197 works over 242 artists for training and evaluating the CNN; (iv) we show that the 3D attributes trained on this dataset generalize to images of other (non-sculpture) object classes; (v) we show that the CNN also provides a shape embedding that can be used to match previously unseen sculptures largely independent of viewpoint; and furthermore (vi) we analyze how the CNN predicts these attributes. Read More

In recent years, we have seen tremendous progress in the field of object detection. Most of the recent improvements have been achieved by targeting deeper feedforward networks. However, many hard object categories, such as bottle and remote, require representation of fine details and not coarse, semantic representations. Read More

Actions are more than just movements and trajectories: we cook to eat and we hold a cup to drink from it. A thorough understanding of videos requires going beyond appearance modeling and necessitates reasoning about the sequence of activities, as well as the higher-level constructs such as intentions. But how do we model and reason about these? We propose a fully-connected temporal CRF model for reasoning over various aspects of activities that includes objects, actions, and intentions, where the potentials are predicted by a deep network. Read More

One characteristic that sets humans apart from modern learning-based computer vision algorithms is the ability to acquire knowledge about the world and use that knowledge to reason about the visual world. Humans can learn about the characteristics of objects and the relationships that occur between them to learn a large variety of visual concepts, often with few examples. This paper investigates the use of structured prior knowledge in the form of knowledge graphs and shows that using this knowledge improves performance on image classification. Read More

There has been a recent paradigm shift in robotics to data-driven learning for planning and control. Due to large number of experiences required for training, most of these approaches use a self-supervised paradigm: using sensors to measure success/failure. However, in most cases, these sensors provide weak supervision at best. Read More

Recently, end-to-end learning frameworks are gaining prevalence in the field of robot control. These frameworks input states/images and directly predict the torques or the action parameters. However, these approaches are often critiqued due to their huge data requirements for learning a task. Read More

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. Convolutional predictors, such as the fully-convolutional network (FCN), have achieved remarkable success by exploiting the spatial redundancy of neighboring pixels through convolutional processing. Though computationally efficient, we point out that such approaches are not statistically efficient during learning precisely because spatial redundancy limits the information learned from neighboring pixels. Read More

Human actions are comprised of a sequence of poses. This makes videos of humans a rich and dense source of human poses. We propose an unsupervised method to learn pose features from videos that exploits a signal which is complementary to appearance and can be used as supervision: motion. Read More

Two less addressed issues of deep reinforcement learning are (1) lack of generalization capability to new target goals, and (2) data inefficiency i.e., the model requires several (and often costly) episodes of trial and error to converge, which makes it impractical to be applied to real-world scenarios. Read More

Large-scale annotated datasets allow AI systems to learn from and build upon the knowledge of the crowd. Many crowdsourcing techniques have been developed for collecting image annotations. These techniques often implicitly rely on the fact that a new input image takes a negligible amount of time to perceive. Read More

In a given scene, humans can often easily predict a set of immediate future events that might happen. However, generalized pixel-level anticipation in computer vision systems is difficult because machine learning struggles with the ambiguity inherent in predicting the future. In this paper, we focus on predicting the dense trajectory of pixels in a scene, specifically what will move in the scene, where it will travel, and how it will deform over the course of one second. Read More

What does a typical visit to Paris look like? Do people first take photos of the Louvre and then the Eiffel Tower? Can we visually model a temporal event like "Paris Vacation" using current frameworks? In this paper, we explore how we can automatically learn the temporal aspects, or storylines of visual concepts from web data. Previous attempts focus on consecutive image-to-image transitions and are unsuccessful at recovering the long-term underlying story. Our novel Skipping Recurrent Neural Network (S-RNN) model does not attempt to predict each and every data point in the sequence, like classic RNNs. Read More

Multi-task learning in Convolutional Networks has displayed remarkable success in the field of recognition. This success can be largely attributed to learning shared representations from multiple supervisory tasks. However, existing multi-task approaches rely on enumerating multiple network architectures specific to the tasks at hand, that do not generalize. Read More

The field of object detection has made significant advances riding on the wave of region-based ConvNets, but their training procedure still includes many heuristics and hyperparameters that are costly to tune. We present a simple yet surprisingly effective online hard example mining (OHEM) algorithm for training region-based ConvNet detectors. Our motivation is the same as it has always been -- detection datasets contain an overwhelming number of easy examples and a small number of hard examples. Read More

We introduce an approach that leverages surface normal predictions, along with appearance cues, to retrieve 3D models for objects depicted in 2D still images from a large CAD object library. Critical to the success of our approach is the ability to recover accurate surface normals for objects in the depicted scene. We introduce a skip-network model built on the pre-trained Oxford VGG convolutional neural network (CNN) for surface normal prediction. Read More

Computer vision has a great potential to help our daily lives by searching for lost keys, watering flowers or reminding us to take a pill. To succeed with such tasks, computer vision methods need to be trained from real and diverse examples of our daily dynamic scenes. While most of such scenes are not particularly exciting, they typically do not appear on YouTube, in movies or TV broadcasts. Read More

What is the right supervisory signal to train visual representations? Current approaches in computer vision use category labels from datasets such as ImageNet to train ConvNets. However, in case of biological agents, visual representation learning does not require millions of semantic labels. We argue that biological agents use physical interactions with the world to learn visual representations unlike current vision systems which just use passive observations (images and videos downloaded from web). Read More

What is a good vector representation of an object? We believe that it should be generative in 3D, in the sense that it can produce new 3D objects; as well as be predictable from 2D, in the sense that it can be perceived from 2D images. We propose a novel architecture, called the TL-embedding network, to learn an embedding space with these properties. The network consists of two components: (a) an autoencoder that ensures the representation is generative; and (b) a convolutional network that ensures the representation is predictable. Read More

What happens if one pushes a cup sitting on a table toward the edge of the table? How about pushing a desk against a wall? In this paper, we study the problem of understanding the movements of objects as a result of applying external forces to them. For a given force vector applied to a specific location in an image, our goal is to predict long-term sequential movements caused by that force. Doing so entails reasoning about scene geometry, objects, their attributes, and the physical rules that govern the movements of objects. Read More

Current generative frameworks use end-to-end learning and generate images by sampling from uniform noise distribution. However, these approaches ignore the most basic principle of image formation: images are product of: (a) Structure: the underlying 3D model; (b) Style: the texture mapped onto structure. In this paper, we factorize the image generation process and propose Style and Structure Generative Adversarial Network (S^2-GAN). Read More

What defines an action like "kicking ball"? We argue that the true meaning of an action lies in the change or transformation an action brings to the environment. In this paper, we propose a novel representation for actions by modeling an action as a transformation which changes the state of the environment before the action happens (precondition) to the state after the action (effect). Motivated by recent advancements of video representation using deep learning, we design a Siamese network which models the action as a transformation on a high-level feature space. Read More

Current learning-based robot grasping approaches exploit human-labeled datasets for training the models. However, there are two problems with such a methodology: (a) since each object can be grasped in multiple ways, manually labeling grasp locations is not a trivial task; (b) human labeling is biased by semantics. While there have been attempts to train robots using trial-and-error experiments, the amount of data used in such experiments remains substantially low and hence makes the learner prone to over-fitting. Read More

This work explores the use of spatial context as a source of free and plentiful supervisory signal for training a rich visual representation. Given only a large, unlabeled image collection, we extract random pairs of patches from each image and train a convolutional neural net to predict the position of the second patch relative to the first. We argue that doing well on this task requires the model to learn to recognize objects and their parts. Read More

We present an approach to utilize large amounts of web data for learning CNNs. Specifically inspired by curriculum learning, we present a two-step approach for CNN training. First, we use easy images to train an initial visual representation. Read More

The field of functional recognition or affordance estimation from images has seen a revival in recent years. As originally proposed by Gibson, the affordances of a scene were directly perceived from the ambient light: in other words, functional properties like sittable were estimated directly from incoming pixels. Recent work, however, has taken a mediated approach in which affordances are derived by first estimating semantics or geometry and then reasoning about the affordances. Read More

Is strong supervision necessary for learning a good visual representation? Do we really need millions of semantically-labeled images to train a Convolutional Neural Network (CNN)? In this paper, we present a simple yet surprisingly powerful approach for unsupervised learning of CNN. Specifically, we use hundreds of thousands of unlabeled videos from the web to learn visual representations. Our key idea is that visual tracking provides the supervision. Read More

Given a scene, what is going to move, and in what direction will it move? Such a question could be considered a non-semantic form of action prediction. In this work, we present a convolutional neural network (CNN) based approach for motion prediction. Given a static image, this CNN predicts the future motion of each and every pixel in the image in terms of optical flow. Read More

Building on the success of recent discriminative mid-level elements, we propose a surprisingly simple approach for object detection which performs comparable to the current state-of-the-art approaches on PASCAL VOC comp-3 detection challenge (no external data). Through extensive experiments and ablation analysis, we show how our approach effectively improves upon the HOG-based pipelines by adding an intermediate mid-level representation for the task of object detection. This representation is easily interpretable and allows us to visualize what our object detector "sees". Read More

Convolutional neural network (CNN) models have demonstrated great success in various computer vision tasks including image classification and object detection. However, some equally important tasks such as visual tracking remain relatively unexplored. We believe that a major hurdle that hinders the application of CNN to visual tracking is the lack of properly labeled training data. Read More

In the past few years, convolutional neural nets (CNN) have shown incredible promise for learning visual representations. In this paper, we use CNNs for the task of predicting surface normals from a single image. But what is the right architecture we should use? We propose to build upon the decades of hard work in 3D scene understanding, to design new CNN architecture for the task of surface normal estimation. Read More

The goal of this paper is to discover a set of discriminative patches which can serve as a fully unsupervised mid-level visual representation. The desired patches need to satisfy two requirements: 1) to be representative, they need to occur frequently enough in the visual world; 2) to be discriminative, they need to be different enough from the rest of the visual world. The patches could correspond to parts, objects, "visual phrases", etc. Read More

In this paper, the generation of topological energy in models with large extra dimensions is investigated. The origin of this energy is attributed to a topological deformation of the standard Minkowski vacuum due to compactification of extra dimensions. This deformation is seen to give rise to an effective, finite energy density due to massive Kaluza-Klein modes of gravitation. Read More

A new mechanism is proposed for generating neutrino masses radiatively through a non-minimal coupling to gravity of fermionic bilinears involving massive neutral fermions. Such coupling terms can arise in theories where the gravity sector is augmented by a scalar field. They necessarily violate the principle of equivalence, but such violations are not ruled out by present experiments. Read More

In the stabilised Randall-Sundrum brane world scenario, the radion can have phenomenologically testable effects, which can be measured against precisely measured electroweak physics data. We investigate the effect of two loop radion corrections to $K_L$ - $K_S$ mass difference to set a bound on the radion mass and vacuum expectation value. It is found that the leading two loop corrections are of the order $[Log(\frac{\Lambda^2}{m_\phi^2}) ]^2$ where $\Lambda$ is the cut-off scale ${\cal O}(\sim$TeV) and $m_\phi$ is the radion mass. Read More

Affiliations: 1HRI, Allahbad, 2Univ. of Delhi, 3Univ. of Delhi, 4Univ. of Delhi

Recently proposed stabilization mechanism of the Randall-Sundrum metric gives rise to a scalar radion, which couples universally to matter with a weak interaction ($\simeq 1$ TeV) scale. Demanding that gauge boson scattering as described by the effective low enerrgy theory be unitary upto a given scale leads to significant constraints on the mass of such a radion. Read More

Affiliations: 1Delhi University, 2Delhi University, 3Delhi University

If the neutrinos are to be identified with the primary source of ultra-high energy cosmic rays(UHECR), their interaction on relic neutrinos is of great importance in understanding their long intergalactic journey. In theories with large compact dimensions, the exchange of a tower of massive spin-2 gravitons (Kaluza-Klein excitations) gives extra contribution to $\nu\bar{\nu} \longrightarrow f\bar{f}$ and $\gamma\gamma$ processes along with the opening of a new channel for the neutrinos to annihilate with the relic cosmic neutrino background $\nu\bar{\nu} \longrightarrow G_{kk}$ to produce bulk gravitons in the extra dimensions. This will affect their attenuation. Read More

A complete experiment on decay $K_L \to l^+ l^-$ will not only consist of measurement of the decay rates but also lepton polarization etc. These additional observations will yield tests of CP invariance in these decays. In $K_L$ and $K_S$ decays, the e mode is slower than the $\mu$ mode by roughly $(m_e/m_\mu)^2$ \cite{sehgal1}. Read More

Rare B decays provides an opportunity to probe for new physics beyond the standard model. the effective Hamiltonian for the decay $b \to s l^+ l^-$ predicts the characteristic polarization for the final state lepton. Lepton polarization has, in addition to a longitudinal component $P_L$, two orthogonal components $P_T$ and $P_N$ lying in and perpendicular to the decay plane. Read More