Ankan Bansal

Haofu Liao, Aruni RoyChowdhury, Weijian Li et al. · amazon-science

We present a new formulation for structured information extraction (SIE) from visually rich documents. It aims to address the limitations of existing IOB tagging or graph-based formulations, which are either overly reliant on the correct ordering of input text or struggle with decoding a complex graph. Instead, motivated by anchor-based object detectors in vision, we represent an entity as an anchor word and a bounding box, and represent entity linking as the association between anchor words. This is more robust to text ordering, and maintains a compact graph for entity linking. The formulation motivates us to introduce 1) a DOCument TRansformer (DocTr) that aims at detecting and associating entity bounding boxes in visually rich documents, and 2) a simple pre-training strategy that helps learn entity detection in the context of language. Evaluations on three SIE benchmarks show the effectiveness of the proposed formulation, and the overall approach outperforms existing solutions.

Shlok Mishra, Anshul Shah, Ankan Bansal et al.

A core component of the recent success of self-supervised learning is cropping data augmentation, which selects sub-regions of an image to be used as positive views in the self-supervised loss. The underlying assumption is that randomly cropped and resized regions of a given image share information about the objects of interest, which the learned representation will capture. This assumption is mostly satisfied in datasets such as ImageNet where there is a large, centered object, which is highly likely to be present in random crops of the full image. However, in other datasets such as OpenImages or COCO, which are more representative of real world uncurated data, there are typically multiple small objects in an image. In this work, we show that self-supervised learning based on the usual random cropping performs poorly on such datasets. We propose replacing one or both of the random crops with crops obtained from an object proposal algorithm. This encourages the model to learn both object and scene level semantic representations. Using this approach, which we call object-aware cropping, results in significant improvements over scene cropping on classification and object detection benchmarks. For example, on OpenImages, our approach achieves an improvement of 8.8% mAP over random scene-level cropping using MoCo-v2 based pre-training. We also show significant improvements on COCO and PASCAL-VOC object detection and segmentation tasks over the state-of-the-art self-supervised learning approaches. Our approach is efficient, simple and general, and can be used in most existing contrastive and non-contrastive self-supervised learning frameworks.

Shlok Mishra, Anshul Shah, Ankan Bansal et al.

Recent literature has shown that features obtained from supervised training of CNNs may over-emphasize texture rather than encoding high-level information. In self-supervised learning in particular, texture as a low-level cue may provide shortcuts that prevent the network from learning higher level representations. To address these problems we propose to use classic methods based on anisotropic diffusion to augment training using images with suppressed texture. This simple method helps retain important edge information and suppress texture at the same time. We empirically show that our method achieves state-of-the-art results on object detection and image classification with eight diverse datasets in either supervised or self-supervised learning tasks such as MoCoV2 and Jigsaw. Our method is particularly effective for transfer learning tasks and we observed improved performance on five standard transfer learning datasets. The large improvements (up to 11.49\%) on the Sketch-ImageNet dataset, DTD dataset and additional visual analyses with saliency maps suggest that our approach helps in learning better representations that better transfer.

Anshul Shah, Shlok Mishra, Ankan Bansal et al.

Recent progress on action recognition has mainly focused on RGB and optical flow features. In this paper, we approach the problem of joint-based action recognition. Unlike other modalities, constellation of joints and their motion generate models with succinct human motion information for activity recognition. We present a new model for joint-based action recognition, which first extracts motion features from each joint separately through a shared motion encoder before performing collective reasoning. Our joint selector module re-weights the joint information to select the most discriminative joints for the task. We also propose a novel joint-contrastive loss that pulls together groups of joint features which convey the same action. We strengthen the joint-based representations by using a geometry-aware data augmentation technique which jitters pose heatmaps while retaining the dynamics of the action. We show large improvements over the current state-of-the-art joint-based approaches on JHMDB, HMDB, Charades, AVA action recognition datasets. A late fusion with RGB and Flow-based approaches yields additional improvements. Our model also outperforms the existing baseline on Mimetics, a dataset with out-of-context actions.

Ankan Bansal, Yuting Zhang, Rama Chellappa

We introduce the task of Image-Set Visual Question Answering (ISVQA), which generalizes the commonly studied single-image VQA problem to multi-image settings. Taking a natural language question and a set of images as input, it aims to answer the question based on the content of the images. The questions can be about objects and relationships in one or more images or about the entire scene depicted by the image set. To enable research in this new topic, we introduce two ISVQA datasets - indoor and outdoor scenes. They simulate the real-world scenarios of indoor image collections and multiple car-mounted cameras, respectively. The indoor-scene dataset contains 91,479 human annotated questions for 48,138 image sets, and the outdoor-scene dataset has 49,617 questions for 12,746 image sets. We analyze the properties of the two datasets, including question-and-answer distributions, types of questions, biases in dataset, and question-image dependencies. We also build new baseline models to investigate new research challenges in ISVQA.

Ankan Bansal, Sai Saketh Rambhatla, Abhinav Shrivastava et al.

The relative spatial layout of a human and an object is an important cue for determining how they interact. However, until now, spatial layout has been used just as side-information for detecting human-object interactions (HOIs). In this paper, we present a method for exploiting this spatial layout information for detecting HOIs in images. The proposed method consists of a layout module which primes a visual module to predict the type of interaction between a human and an object. The visual and layout modules share information through lateral connections at several stages. The model uses predictions from the layout module as a prior to the visual module and the prediction from the visual module is given as the final output. It also incorporates semantic information about the object using word2vec vectors. The proposed model reaches an mAP of 24.79% for HICO-Det dataset which is about 2.8% absolute points higher than the current state-of-the-art.

Prithviraj Dhar, Ankan Bansal, Carlos D. Castillo et al.

As deep networks become increasingly accurate at recognizing faces, it is vital to understand how these networks process faces. While these networks are solely trained to recognize identities, they also contain face related information such as sex, age, and pose of the face. The networks are not trained to learn these attributes. We introduce expressivity as a measure of how much a feature vector informs us about an attribute, where a feature vector can be from internal or final layers of a network. Expressivity is computed by a second neural network whose inputs are features and attributes. The output of the second neural network approximates the mutual information between feature vectors and an attribute. We investigate the expressivity for two different deep convolutional neural network (DCNN) architectures: a Resnet-101 and an Inception Resnet v2. In the final fully connected layer of the networks, we found the order of expressivity for facial attributes to be Age > Sex > Yaw. Additionally, we studied the changes in the encoding of facial attributes over training iterations. We found that as training progresses, expressivities of yaw, sex, and age decrease. Our technique can be a tool for investigating the sources of bias in a network and a step towards explaining the network's identity decisions.

Ankan Bansal, Sai Saketh Rambhatla, Abhinav Shrivastava et al.

We present an approach for detecting human-object interactions (HOIs) in images, based on the idea that humans interact with functionally similar objects in a similar manner. The proposed model is simple and efficiently uses the data, visual features of the human, relative spatial orientation of the human and the object, and the knowledge that functionally similar objects take part in similar interactions with humans. We provide extensive experimental validation for our approach and demonstrate state-of-the-art results for HOI detection. On the HICO-Det dataset our method achieves a gain of over 2.5% absolute points in mean average precision (mAP) over state-of-the-art. We also show that our approach leads to significant performance gains for zero-shot HOI detection in the seen object setting. We further demonstrate that using a generic object detector, our model can generalize to interactions involving previously unseen objects.

Rajeev Ranjan, Ankan Bansal, Jingxiao Zheng et al.

The availability of large annotated datasets and affordable computation power have led to impressive improvements in the performance of CNNs on various object detection and recognition benchmarks. These, along with a better understanding of deep learning methods, have also led to improved capabilities of machine understanding of faces. CNNs are able to detect faces, locate facial landmarks, estimate pose, and recognize faces in unconstrained images and videos. In this paper, we describe the details of a deep learning pipeline for unconstrained face identification and verification which achieves state-of-the-art performance on several benchmark datasets. We propose a novel face detector, Deep Pyramid Single Shot Face Detector (DPSSD), which is fast and capable of detecting faces with large scale variations (especially tiny faces). We give design details of the various modules involved in automatic face recognition: face detection, landmark localization and alignment, and face identification/verification. We provide evaluation results of the proposed face detector on challenging unconstrained face detection datasets. Then, we present experimental results for IARPA Janus Benchmarks A, B and C (IJB-A, IJB-B, IJB-C), and the Janus Challenge Set 5 (CS5).

Ankan Bansal, Karan Sikka, Gaurav Sharma et al.

We introduce and tackle the problem of zero-shot object detection (ZSD), which aims to detect object classes which are not observed during training. We work with a challenging set of object classes, not restricting ourselves to similar and/or fine-grained categories as in prior works on zero-shot classification. We present a principled approach by first adapting visual-semantic embeddings for ZSD. We then discuss the problems associated with selecting a background class and motivate two background-aware approaches for learning robust detectors. One of these models uses a fixed background class and the other is based on iterative latent assignments. We also outline the challenge associated with using a limited number of training classes and propose a solution based on dense sampling of the semantic label space using auxiliary data with a large number of categories. We propose novel splits of two standard detection datasets - MSCOCO and VisualGenome, and present extensive empirical results in both the traditional and generalized zero-shot settings to highlight the benefits of the proposed methods. We provide useful insights into the algorithm and conclude by posing some open questions to encourage further research.

Rajeev Ranjan, Ankan Bansal, Hongyu Xu et al.

In recent years, the performance of face verification and recognition systems based on deep convolutional neural networks (DCNNs) has significantly improved. A typical pipeline for face verification includes training a deep network for subject classification with softmax loss, using the penultimate layer output as the feature descriptor, and generating a cosine similarity score given a pair of face images or videos. The softmax loss function does not optimize the features to have higher similarity score for positive pairs and lower similarity score for negative pairs, which leads to a performance gap. In this paper, we propose a new loss function, called Crystal Loss, that restricts the features to lie on a hypersphere of a fixed radius. The loss can be easily implemented using existing deep learning frameworks. We show that integrating this simple step in the training pipeline significantly improves the performance of face verification and recognition systems. We achieve state-of-the-art performance for face verification and recognition on challenging LFW, IJB-A, IJB-B and IJB-C datasets over a large range of false alarm rates (10-1 to 10-7).

Sayantan Sarkar, Ankan Bansal, Upal Mahbub et al.

In this paper, targeted fooling of high performance image classifiers is achieved by developing two novel attack methods. The first method generates universal perturbations for target classes and the second generates image specific perturbations. Extensive experiments are conducted on MNIST and CIFAR10 datasets to provide insights about the proposed algorithms and show their effectiveness.

Ankan Bansal, Carlos Castillo, Rajeev Ranjan et al.

While the research community appears to have developed a consensus on the methods of acquiring annotated data, design and training of CNNs, many questions still remain to be answered. In this paper, we explore the following questions that are critical to face recognition research: (i) Can we train on still images and expect the systems to work on videos? (ii) Are deeper datasets better than wider datasets? (iii) Does adding label noise lead to improvement in performance of deep networks? (iv) Is alignment needed for face recognition? We address these questions by training CNNs using CASIA-WebFace, UMDFaces, and a new video dataset and testing on YouTube- Faces, IJB-A and a disjoint portion of UMDFaces datasets. Our new data set, which will be made publicly available, has 22,075 videos and 3,735,476 human annotated frames extracted from them.

Ankan Bansal, Anirudh Nanduri, Carlos Castillo et al.

Recent progress in face detection (including keypoint detection), and recognition is mainly being driven by (i) deeper convolutional neural network architectures, and (ii) larger datasets. However, most of the large datasets are maintained by private companies and are not publicly available. The academic computer vision community needs larger and more varied datasets to make further progress. In this paper we introduce a new face dataset, called UMDFaces, which has 367,888 annotated faces of 8,277 subjects. We also introduce a new face recognition evaluation protocol which will help advance the state-of-the-art in this area. We discuss how a large dataset can be collected and annotated using human annotators and deep networks. We provide human curated bounding boxes for faces. We also provide estimated pose (roll, pitch and yaw), locations of twenty-one key-points and gender information generated by a pre-trained neural network. In addition, the quality of keypoint annotations has been verified by humans for about 115,000 images. Finally, we compare the quality of the dataset with other publicly available face datasets at similar scales.

Ankan Bansal, K. S. Venkatesh

We present a method of estimating the number of people in high density crowds from still images. The method estimates counts by fusing information from multiple sources. Most of the existing work on crowd counting deals with very small crowds (tens of individuals) and use temporal information from videos. Our method uses only still images to estimate the counts in high density images (hundreds to thousands of individuals). At this scale, we cannot rely on only one set of features for count estimation. We, therefore, use multiple sources, viz. interest points (SIFT), Fourier analysis, wavelet decomposition, GLCM features and low confidence head detections, to estimate the counts. Each of these sources gives a separate estimate of the count along with confidences and other statistical measures which are then combined to obtain the final estimate. We test our method on an existing dataset of fifty images containing over 64000 individuals. Further, we added another fifty annotated images of crowds and tested on the complete dataset of hundred images containing over 87000 individuals. The counts per image range from 81 to 4633. We report the performance in terms of mean absolute error, which is a measure of accuracy of the method, and mean normalised absolute error, which is a measure of the robustness.