Virtual World Creation

Aryamaan Jain

Abstract

Synthesis, capture and analysis of a highly complex 3D terrain structure are essential for critical applications such as river/flood modelling, disaster mitigation planning, landslide modelling and flight simulation. On the other hand, synthesis of natural-looking 3D terrains finds its applications in the entertainment industry such as computer gaming and VFX. This thesis explores novel learning-based techniques for the generation of immersive and realistic 3D virtual environments, catering to the needs of the aforementioned applications. The generation of virtual worlds involves multiple components, including terrain, vegetation, and other objects. We primarily focus on three key aspects for virtual world generation: 1) develop novel AI-enabled 3D terrain authoring solutions based on real-world satellite and aerial data using a learning-based framework, 2) L-systems grammar-based 3D tree generation and 3) rendering techniques that are used to create high-quality visualizations of the generated world. Terrain generation is a critical component of 3D virtual world generation, as it provides the foundational structure for the environment. Traditional techniques for 3D terrain generation involve procedural generation, which relies on mathematical algorithms to generate landscapes. However, deep learning techniques have shown promise in generating more realistic terrain, as they can learn from real-world data to produce new, varied, and realistic landscapes. In this thesis, we explore the use of deep learning techniques for 3D terrain generation, which can produce realistic and varied terrains with high visual fidelity. Specifically, we propose two learning-based novel frameworks for Interactive 3D Terrain Authoring & Manipulation and Adaptive Multi-Resolution Infinite Terrain Generation. In addition to terrain generation, vegetation is another important component of virtual world generation. Trees and other plants provide visual interest and can help create a more immersive environment. L-systems are a popular technique for generating realistic vegetation, as they are capable of generating complex structures that resemble real-world plants. In this thesis, we propose a variant of the L-systems for 3D tree generation and compare the results to traditional procedural generation techniques. Finally, rendering is a critical component of 3D virtual world generation, as it is responsible for creating the final visual output that users will see. In addition to terrain and tree generation, this thesis also covers rendering techniques used to visualize the generated virtual world. We explore the use of real-time rendering techniques in conjunction with terrain generation to achieve high-quality visual results while maintaining performance. Overall, the research presented in this thesis aims to advance the state-of-the-art in virtual world generation and contribute to the development of more realistic and immersive virtual environments. We performed extensive empirical evaluation on publicly available datasets to report details qualitative and quantitative results demonstrating the superiority of the proposed methods over existing solutions in the literature.

Year of completion:  July 2023
 Advisor : Avinash Sharma, Rajan Krishnan Sundara

Related Publications

    Downloads

    thesis

    Driving into the Dataverse: Real and Synthetic Data for Autonomous Vehicles

    Shubham Dokania

    Abstract

    The rapid advancement of autonomous driving systems is transforming the future of transportation and urban mobility. However, these systems face significant challenges when deployed in complex and unstructured traffic environments, such as those found in many cities in south-east Asian countries like India. This thesis aims to address the challenges related to data collection, management, and generation for autonomous driving systems operating in such scenarios. The primary contributions of this work include the development of a data collection toolkit, the creation of a comprehensive driving dataset for unstructured environments (IDD-3D), and the proposal of a synthetic data generation framework (TRoVE) based on real-world data. The data collection toolkit presented in this thesis enables sensor fusion for driving scenarios by treating sensor APIs as separate entities from the data collection interface. This framework allows for the use of any sensor configuration and demonstrates the smooth creation of driving datasets using our framework. This toolkit is adaptable to different environments and can be easily scaled, making it a crucial step towards creating a large-scale dataset for Indian road scenarios. The IDD-3D dataset provides a valuable resource for studying unstructured driving scenarios with complex road situations. We present a thorough statistical and experimental analysis of the dataset, which includes high-quality annotations for 3D object bounding boxes and instance IDs for tracking. We highlight the diverse object types, categories, and complex trajectories found in Indian road scenes, enabling the development of robust autonomous driving systems that can generalize across different geographical locations. In addition, we provide benchmarks for 3D object detection and tracking using state-of-the-art approaches. The TRoVE synthetic data toolkit offers a framework for the automatic generation of synthetic data for visual perception, leveraging existing real-world data. By combining synthetic data with real data, we show the potential for improved performance in various computer vision tasks. The data generation process can be extended to different locations and scenarios, avoiding the limitations of bounded data volumes and variety found in manually designed virtual environments. This thesis contributes to the development of systems in complex environments by addressing the challenges of data acquisition, management, and generation. By bridging the gap between the current state-of-the-art and the needs of unstructured traffic scenarios, this work paves the way for more robust and versatile intelligent transportation systems that can operate safely and efficiently in a wide range of situations

    Year of completion:  July 2023
     Advisor : C V Jawahar

    Related Publications

      Downloads

      thesis

      Improved Representation Spaces for Videos

      Bipasha Sen

      Abstract

      Videos form an integral part of human lives and act as one of the most natural forms of perception spanning both the spatial and the temporal dimensions: the spatial dimension emphasizes the content, whereas the temporal dimension emphasizes the change. Naturally, studying this modality is an important area of computer vision. Notably, one must efficiently capture this high-dimensional modality to perform different downstream tasks robustly. In this thesis, we study representation learning for videos to perform two key aspects of video-based tasks: classification and generation. In a classification task, a video is compressed to a latent space that captures the key discriminative properties of a video relevant to the task. On the other hand, generation involves starting with a latent space (often a known space, such as standard normal) and learning a valid mapping between the latent and the video manifold. This thesis explores complementary representation techniques to develop robust representation spaces useful for diverse downstream tasks. In this vein, this thesis starts by tackling video classification, where we concentrate on a specific task of “lipreading” (transliterating videos to text) or in technical terms - classifying videos of mouth movements. Through this work, we propose a compressed generative space that self-augments the dataset improving the discriminative capabilities of the classifier. Motivated by the findings of this work, we move on to finding an improved generative space in which we touch upon several key elements of video generation, including unconditional video generation, video inversion, and video superresolution. In the classification task, we aim to study lipreading (or visually recognizing speech from the mouth movements of a speaker), a challenging and mentally taxing task for humans to perform. Unfortunately, multiple medical conditions force people to depend on this skill in their day-to-day lives for essential communication. Patients suffering from ‘Amyotrophic Lateral Sclerosis’ (ALS) often lose muscle control, consequently, their ability to generate speech and communicate via lip movements. Existing large datasets do not focus on medical patients or curate personalized vocabulary relevant to an individual. Collecting large-scale datasets of a patient needed to train modern data-hungry deep learning models is, however, extremely challenging. We propose a personalized network designed to lipread for an ALS patient using only one-shot examples. We depend on synthetically generated lip movements to augment the one-shot scenario. A Variational Encoder-based domain adaptation technique is used to bridge the real-synthetic domain gap. Our approach significantly improves and achieves high top-5 accuracy with 83.2% accuracy compared to 62.6% achieved by comparable methods for the patient. Apart from evaluating our approach on the ALS patient, we also extend it to people with hearing impairment, relying extensively on lip movements to communicate. In the next part of the thesis, we focus on representation spaces for video-based generative tasks. Generating videos is a complex task that is accomplished by generating a set of temporally coherent images frame-by-frame. This approach confines the expressivity of videos to image-based operations on individual frames, necessitating network designs that can achieve temporally coherent trajectories in the underlying image space. We propose INR-V, a video representation network that learns a continuous space for video-based generative tasks. INR-V parameterizes videos using implicit neural representations (INRs), a multi-layered perceptron that predicts an RGB value for each input pixel location of the video. The INR is predicted using a meta-network which is a hypernetwork trained on neural representations of multiple video instances. Later, the meta-network can be sampled to generate diverse novel videos enabling many downstream video-based generative tasks. Interestingly, we find that conditional regularization and progressive weight initialization play a crucial role in obtaining INR-V. The representation space learned by INR-V is more expressive than an image space showcasing many interesting properties not possible with the existing works. For instance, INR-V can smoothly interpolate intermediate videos between known video instances (such as intermediate identities, expressions, and poses in face videos). It can also in-paint missing portions in videos to recover temporally coherent full videos. We evaluate the space learned by INR-V on diverse generative tasks such as video interpolation, novel video generation, video inversion, and video inpainting against the existing baselines. INR-V significantly outperforms the baselines on several of these demonstrated tasks, clearly showcasing the potential of the proposed representation space. In summary, this thesis makes a significant contribution to the field of computer vision by exploring representation learning for videos. The proposed methods are thoroughly evaluated through extensive experimentation and analysis, which clearly demonstrate their advantages over existing works. These findings have the potential to advance a range of video-based applications, including personalized healthcare, entertainment, and communication. By developing robust representation spaces that improve video classification and generation, this work opens up new possibilities for more natural and effective ways of perceiving, understanding, and interacting with videos

      Year of completion:  August 2023
       Advisor : C V Jawahar, Vinay P Namboodiri

      Related Publications

        Downloads

        thesis

        Open-Vocabulary Audio Keyword Spotting with Low Resource Language Adaptation

        Kirandevraj R

        Abstract

        Open-Vocabulary Keyword spotting solves the problem of spotting audio keywords in an utterance. The keyword set can incorporate keywords the system has seen and not seen during training, functioning as zero-shot keyword spotting. The traditional method involves using ASR to transcribe audio to text and search in the text space to spot keywords. Other methods include obtaining the posteriors from a Deep Neural Network (DNN) and using template matching algorithms to find similarities. Keyword spotting does not require transcribing the entire audio and focuses on detecting the specific words of interest. In this thesis, we aim to explore the usage of the Automatic Speech Recognition (ASR) system for Keyword Spotting. We demonstrate that the intermediate representation of ASR can be used for open vocabulary keyword spotting. With this, we show the effectiveness of using Connectionist Temporal Classification (CTC) loss for learning word embeddings for keyword spotting. We propose a novel method of using the CTC loss function with the traditional triplet loss function to learn word embeddings for keyword spotting on the TIMIT English language audio dataset. We show this method achieves an Average Precision (AP) of 0.843 over 344 words unseen by the model trained on the TIMIT dataset. In contrast, the Multi-View recurrent method that learns jointly on the text and acoustic embeddings achieves only 0.218 for out-of-vocabulary words. We propose a novel method to generalize our approach to Tamil, Vallader, and Hausa low-resource languages. Here we use transliteration to convert the Tamil language script to English such that the Tamil words sound similar written with English alphabets. The model predicts the transliterated text for input Tamil audio with CTC and triplet loss functions. We show that this method helps transfer the knowledge learned from high resource language English to low resource language Tamil. We further reduce the model size to make it work in a small footprint scenario like mobile phones. To this extent, we explore various knowledge distillation loss functions such as MSE, KL Divergence, and CosEmbedding loss functions. We observe that small-footprint ASR representation is competitive with knowledge distillation methods for small-footprint keyword spotting. This methodology makes use of existing ASR networks trained with massive datasets. It converts them into open vocabulary keyword spotting systems that can also be generalized to low-resource language.

        Year of completion:  November 2022
         Advisor : C V Jawahar,Vinay P Namboodiri,Vinod kumar Kurmi

        Related Publications

          Downloads

          thesis

          On Designing Efficient Deep Neural Networks for Semantic Segmentation

          Nikitha Vallurupalli

          Abstract

          Semantic segmentation is an essential primitive in real-time systems such as autonomous navigation, which require processing at high frames per second. Hence for models to be practically applicable, it is essential that they have to be compact, fast as well as achieve high prediction accuracies. Previous research into semantic segmentation has focused on creating high-performance deep learning architectures. Most of the time, these best-performing models are complex, deep, have large processing times, and demand a significantly higher amount of processing capacity. Another relevant area of research is model compression, by which we can obtain lightweight models. Considering that there also have been works that produced mainstream light-weight semantic segmentation models at the expense of performance, we design models that bring a desirable balance between performance and latency. Specifically, methods and architectures that give a high performance while being real-time and working on resourceconstrained settings. We identify the redundancies in the existing state-of-the-art approaches and propose compact architecture family called ESSNet with accuracy comparable to the state-of-the-art while utilizing only a fraction of the space and computational power of those networks. We propose convolutional module designs with sparse coding theory as a premise and we also present two real-time encoder backbones employing our proposed modules. We empirically evaluate the efficacy of our proposed layers and compare them with existing approaches. Secondly, we explore the need for optimization during the training phase in the proposed models and present a novel training method called Gradual Grouping that results in models with improved implementation efficiency vs accuracy trade-offs. Additionally, we conduct extensive experiments by varying architecture hyper-parameters such as network depth, kernel sizes, dilation rates, split branching and additional context extraction modules. We also present a compact architecture using multi-branch separable convolutional layers with different dilation rates.

          Year of completion:  November 2022
           Advisor : C V Jawahar,Girish Varma

          Related Publications

            Downloads

            thesis

            More Articles …

            1. Towards Handwriting Recognition and Search in Indic & Latin Scripts
            2. Learnable HMD Facial De-occlusion for VR Applications in a Person-Specific Setting
            3. Interactive Video Editing using Machine Learning Techniques
            4. Refining 3D Human Digitization Using Learnable Priors