Advances in Robot Perception and Communication Towards Networked Autonomy

Rajat Talak

Date : 29/05/2024

Abstract:

Networked autonomy aspires to build a team of robots that can seamlessly perceive, communicate, and interact with humans and the environment. A robot perception system needs to build an actionable 3D scene representation in real-time that is scalable for long-horizon networked autonomy, and also implement mechanisms that can tackle failures due to changing environmental conditions. Operating in constantly changing environments, networked autonomous systems need to exchange time-sensitive information to ensure situational awareness and safety.

Becoming friends with pixels through intermediate representations

Dr. Kuldeep Kulkarni

Date : 30/08/2023

Abstract:

Manipulation of natural images for tasks like object insertion, out-painting or creating animations is extremely difficult if we operate purely in the pixel domain. Dr Kuldeep Kulkarni talked about the advantages of manipulating visual data by directly expressing them in intermediate representations and manipulating them instead of the pixels. Specifically, the focus was on his recent works with image out-painting and animating still images as target applications. He first talked about a semantically-aware novel paradigm to perform image extrapolation that enables the addition of new object instances. Expressing the images in semantic label space allows us to complete the existing objects more effectively and add completely new objects that otherwise are very difficult when working in the pixel domain. Dr. Kulkarni also discussed methods he developed to exploit intermediate representations like optical flow and surface normal maps to generate cinema graphs depicting the animation of fluid elements and human clothing.

Automatically Generating Audio Descriptions for Movies

Professor Andrew Zisserman

Date : 21/08/2023

Abstract:

Audio Description is the task of generating descriptions of visual content, at suitable time intervals, for the benefit of visually impaired audiences. For movies, this presents notable challenges - the Audio Description must occur only during existing pauses in dialogue, should refer to characters by name, and ought to aid understanding of the storyline as a whole. This requires a visual-language model that can address all three of the `what', `who', and `when' questions: What is happening in the scene? Who are the characters in the scene? And when should a description be given?

Professor Andrew Zisserman visited IIIT-H and gave a talk on the 21st of August, 2023. He discussed how to build on large pre-trained models to construct a visual-language model that can generate Audio Descriptions addressing the following questions: (i) how to incorporate visual information into a pre-trained language model; (ii) how to train the model using only partial information; (iii) how to use a `character bank' to provide information on who is in a scene; and (iv) how to improve the temporal alignment of an ASR model to obtain clean data for training.

Towards Trustworthy and Fair Medical Image Analysis Models

Raghav Mehta

Date : 15/05/2023

Abstract:

Although Deep Learning (DL) models have been shown to perform very well on various medical imaging tasks, inference in the presence of pathology presents several challenges to common models. These challenges impede the integration of DL models into real clinical workflows. Deployment of these models into real clinical contexts requires: (1) that the confidence in DL model predictions be accurately expressed in the form of uncertainties and (2) that they exhibit robustness and fairness across different sub-populations. In this talk, we will look at our recent work, where we developed an uncertainty quantification score for the task of Brain Tumour Segmentation. We evaluated the score's usefulness during the two consecutive Brain Tumour Segmentation (BraTS) challenges, BraTS 2019 and BraTS 2020. Overall, our findings confirm the importance and complementary value that uncertainty estimates provide to segmentation algorithms, highlighting the need for uncertainty quantification in medical image analyses. Additionally, we combine the aspect of uncertainty estimates with fairness across demographic subgroups into the picture. By performing extensive experiments on multiple tasks, we show that popular ML methods for achieving fairness across different subgroups, such as data-balancing and distributionally robust optimization, succeed in terms of the model performances for some of the tasks. However, this can come at the cost of poor uncertainty estimates associated with the model predictions. At last, we talk about our ongoing work on fairness mitigation framework in terms of calibration. Although several methods have been shown to successfully mitigate biases across subgroups in terms of accuracy, they do not consider calibration across different subgroups of these models. To this end, we propose a novel two-stage method Cluster-Focal. Extensive experiments on two different medical image classification datasets show that our method effectively controls calibration error in the worst-performing subgroups while preserving prediction performance, outperforming recent baselines.

Computer vision - Self-Supervised Representation Learning

Yash Patel

Date : 16/03/2023

Abstract:

Yash Patel during his presentation, discussed a range of topics related to his research interests, which primarily include Self-Supervised Representation Learning, Image Compression, Scene Text Detection and Recognition, Tracking and Segmentation in Videos, and 3D Reconstruction.
Specifically, he discussed the challenges associated with Training Neural Networks on Non-Differentiable Losses and presented various approaches for overcoming these challenges. In particular, he presented his proposed technique for training a neural network by minimizing a surrogate loss that approximates a target evaluation metric that may be non-differentiable. To achieve this, the surrogate is learned via a deep embedding method where the Euclidean distance between the prediction and the ground truth corresponds to the value of the evaluation metric. Additionally, he described his work on proposing a differentiable surrogate loss for the recall metric. To enable training with a very large batch size, which is crucial for metrics computed on the entire retrieval database, the speaker utilized an implementation that sidesteps the hardware constraints of the GPU memory. Furthermore, an efficient mixup approach that operates on pairwise scalar similarities was employed to virtually increase the batch size further.

Building Maximal Vision Systems with Minimal Resources

Dr. Ayush Bansal

Date : 19/12/2022

Abstract:

Current vision and robotic systems are like mainframe machines of the 60s -- they require extensive resources: (1) dense data capture and massive human annotations, (2) large parametric models, and (3) intensive computational infrastructure. I build systems that can learn directly from sparse and unconstrained real-world samples with minimal resources, i.e., limited or no supervision, use simple and efficient models, and operate on every day computational devices. Building systems with minimal resources allows us to democratize them for non-experts. My work has impacted important areas such as virtual reality, content creation and audio-visual editing, and providing a natural voice to speech-impaired individuals.

In my talk, I will present my efforts to build vision systems for novel view synthesis. I will discuss Neural Pixel Composition, a novel approach for continuous 3D-4D view synthesis that reliably operates on sparse and wide-baseline multi-view images/videos and can be trained efficiently within a few minutes for high-resolution (12MP) content using 1 GB GPU memory. I will present my efforts to build vision systems for unsupervised audio-visual synthesis. I will primarily discuss Exemplar Autoencoders that enable zero-shot audio-visual retargeting. Exemplar Autoencoders are built on remarkably simple insights: (1) autoencoders project out-of-sample data onto the distribution of the training set; and (2) exemplar learning enables us to capture the voice, stylistic prosody (emotions and ambiance), and visual appearance of the target. These properties enable an autoencoder trained on an individual's voice to generalize for unknown voices in different languages. Exemplar Autoencoders can synthesize natural voices for speech-impaired individuals and do a zero-shot multilingual translation.

Towards Autonomous Driving in Dense, Heterogeneous, and Unstructured Environments

Dr. Rohan Chandra

Date : 07/11/2022

Abstract:

In this talk, I discuss many key problems in autonomous driving towards handling dense, heterogeneous, and unstructured traffic environments. Autonomous vehicles (AV) at present are restricted to operating on smooth and well-marked roads, in sparse traffic, and among well-behaved drivers. I present new techniques to perceive, predict, and navigate among human drivers in traffic that is significantly denser in terms of a number of traffic-agents, more heterogeneous in terms of size and dynamic constraints of traffic agents, and where many drivers may not follow the traffic rules and have varying behaviors. My talk is structured along three themes—perception, driver behavior modeling, and planning. More specifically, I will talk about Improved tracking and trajectory prediction algorithms for dense and heterogeneous traffic using a combination of computer vision and deep learning techniques. A novel behavior modeling approach using graph theory for characterizing human drivers as aggressive or conservative from their trajectories. Behavior-driven planning and navigation algorithms in mixed and unstructured traffic environments using game theory and risk-aware planning. Finally, I will conclude by discussing the future implications and broader applications of these ideas in the context of social robotics where robots are deployed in warehouses, restaurants, hospitals, and inside homes to assist human beings.

Unsupervised Representation Learning

Maneesh Kumar Singh

Date : 28/11/2018

Abstract:

upervised machine learning using deep neural networks have shown tremendous success on a variety of tasks in machine learning. However, supervised learning on each individual task is neither scalable nor the only way to create world models of visual (and other) phenomena and to do inference on them. The recent research thrusts are in mitigating and surmounting such problems. The learning of static world models is also called representation learning. Unsupervised representation learning seeks to create structured latent representations to avoid the onerous need to generate supervisory labels and to enable learning of task-independent (universal) representations. In this talk, I will provide an overview of recent efforts in my group at Verisk, carried out in collaboration with various academic partners, on these topics. Most of this work is available in recent publications and accompanying code online.

Visual recognition of human communications

Dr Joon Son Chung

Date : 05/09/2018

Abstract:

The objective of this work is visual recognition of human communications.Solving this problem opens up a host of applications, such as transcribing archival silent films, or resolving multi-talker simultaneous speech,but most importantly it helps to advance the state of the art in speech recognition by enabling machines to take advantage of the multi-modal nature of human communications.
Training a deep learning algorithm requires a lot of training data.We propose a method to automatically collect, process and generate a large-scale audio-visual corpus from television videos temporally aligned with the transcript.To build such data-set, it is essential to know 'who' is speaking 'when'.We develop a ConvNet model that learns joint embedding of the sound and the mouth images from unlabeled data, and apply this network to the tasks of audio-to-video synchronization and active speaker detection.We also show that the methods developed here can be extended to the problem of generating talking faces from audio and still images, and re-dubbing videos with audio samples from different speakers.
We then propose a number of deep learning models that are able to recognize visual speech at sentence level.The lip reading performance beats a professional lip reader on videos from BBC television.We demonstrate that if audio is available, then visual information helps to improve speech recognition performance.We also propose methods to enhance noisy audio and to resolve multi-talker simultaneous speech using visual cues.
Finally, we explore the problem of speaker recognition.Whereas previous works for speaker identification have been limited to constrained conditions, here we build a new large-scale speaker recognition data-set collected from 'in the wild' videos using an automated pipeline. We propose a number of ConvNet architectures that outperforms traditional baselines on this data-set.

Visual recognition of human communications

Triantafyllos Afouras

Date : 05/09/2018

Abstract:

The objective of this work is visual recognition of human communications.Solving this problem opens up a host of applications, such as transcribing archival silent films, or resolving multi-talker simultaneous speech,but most importantly it helps to advance the state of the art in speech recognition by enabling machines to take advantage of the multi-modal nature of human communications.
Training a deep learning algorithm requires a lot of training data.We propose a method to automatically collect, process and generate a large-scale audio-visual corpus from television videos temporally aligned with the transcript.To build such data-set, it is essential to know 'who' is speaking 'when'.We develop a ConvNet model that learns joint embedding of the sound and the mouth images from unlabeled data, and apply this network to the tasks of audio-to-video synchronization and active speaker detection.We also show that the methods developed here can be extended to the problem of generating talking faces from audio and still images, and re-dubbing videos with audio samples from different speakers.
We then propose a number of deep learning models that are able to recognize visual speech at sentence level.The lip reading performance beats a professional lip reader on videos from BBC television.We demonstrate that if audio is available, then visual information helps to improve speech recognition performance.We also propose methods to enhance noisy audio and to resolve multi-talker simultaneous speech using visual cues.
Finally, we explore the problem of speaker recognition.Whereas previous works for speaker identification have been limited to constrained conditions, here we build a new large-scale speaker recognition data-set collected from 'in the wild' videos using an automated pipeline. We propose a number of ConvNet architectures that outperforms traditional baselines on this data-set.

Making sense of Urban Data: Experiences from interdisciplinary research studies

Dr Manik Gupta

Date : 03/08/2018

Abstract:

In this talk, I present an overview about the different research projects that I have worked during my research career beginning with my PhD on air pollution monitoring to my first postdoc on wheelchair accessibility and then onto my second postdoc on building energy management data to my current successful research grants on ear protection in noisy environments.
Finally, I summarize my learnings and future directions, guiding towards unlimited opportunities to make difference to millions of lives in this very exciting new era of data and more data.

Semantic Representation and Analysis of E-commerce Orders

Dr.Arijit Biswas

University of Maryland

Date : 29/01/2018

Abstract:

E-commerce websites such as Amazon, Alibaba, and Walmart typically process billions of orders every year. Semantic representation and understanding of these orders is extremely critical for an eCommerce company. Each order can be represented as a tuple of <customer, product, price, date>. In this talk, I will describe two of our recent work (i) product embedding using MRNet-Product2Vec and (ii) generating fake orders using eCommerceGAN.

MRNet-Product2Vec [ECML-PKDD 2017]: In this work, we propose an approach called MRNet-Product2Vec for creating generic embeddings of products within an e-commerce ecosystem. We learn a dense and low-dimensional embedding where a diverse set of signals related to a product are explicitly injected into its representation. We train a Discriminative Multi-task Bidirectional Recurrent Neural Network (RNN), where the input is a product title fed through a Bidirectional RNN and at the output, product labels corresponding to fifteen different tasks are predicted.

eCommerceGAN: Exploring the space of all plausible orders could help us better understand the relationships between the various entities in an e-commerce ecosystem, namely the customers and the products they purchase. In this paper, we propose a Generative Adversarial Network (GAN) for orders made in e-commerce websites. Once trained, the generator in the GAN could generate any number of plausible orders. Our contributions include: (a) creating a dense and low-dimensional representation of e-commerce orders, (b) train an ecommerceGAN (ecGAN) with real orders to show the feasibility of the proposed paradigm, and (c) train an ecommerce-conditional-GAN (ec^2GAN) to generate the plausible orders involving a particular product. We propose several qualitative methods to evaluate ecGAN and demonstrate its effectiveness.