Nagar, Rajendra

Implicit 3D surface reconstruction of an object from its partial and noisy 3D point cloud scan is the classical geometry processing and 3D computer vision problem. In the literature, various 3D shape representations have been developed, differing in memory efficiency and shape retrieval effectiveness, such as volumetric, parametric, and implicit surfaces. Radial basis functions provide memory-efficient parameterization of the implicit surface. However, we show that training a neural network using the mean squared error between the groundtruth implicit surface and the linear basis-based implicit surfaces does not converge to the global solution. In this work, we propose locally supported compact radial basis functions for a linear representation of the implicit surface. This representation enables us to generate 3D shapes with arbitrary topologies at any resolution due to their continuous nature. We then propose a neural network architecture for learning the linear implicit shape representation of the 3D surface of an object. We learn linear implicit shapes within a supervised learning framework using ground truth Signed-Distance Field (SDF) data for guidance. The classical strategies face difficulties in finding linear implicit shapes from a given 3D point cloud due to numerical issues (requires solving inverse of a large matrix) in basis and query point selection. The proposed approach achieves better Chamfer distance and comparable F-score than the state-of-the-art approach on the benchmark dataset. We also show the effectiveness of the proposed approach by using it for the 3D shape completion task.

Detecting the reflection symmetry plane of an object represented by a 3D point cloud is a fundamental problem in 3D computer vision and geometry processing due to its various applications, such as compression, object detection, robotic grasping, 3D surface reconstruction, etc. Several approaches exist to solve this problem for clean 3D point clouds. However, it is a challenging problem to solve in the presence of outliers and missing parts. The existing methods try to overcome this challenge primarily by voting-based techniques but do not work efficiently. In this work, we proposed a statistical estimator-based approach for the plane of reflection symmetry that is robust to outliers and missing parts. We pose the problem of finding the optimal estimator for the reflection symmetry as an optimization problem on a 2-sphere that quickly converges to the global solution for an approximate initialization. We further adapt the heat kernel signature for symmetry invariant matching of mirror symmetric points. This approach helps us to decouple the chicken-and-egg problem of finding the optimal symmetry plane and correspondences between the reflective symmetric points. The proposed approach achieves comparable mean ground-truth error and 4.5% increment in the F-score as compared to the state-of-the-art approaches on the benchmark dataset.

The increased satellite launches have made the capture of debris and On-Orbit servicing of the orbiting satellites essential. In space, objects exhibit a tumbling motion around their major inertial axis. In this paper, we propose a featureless approach for a robotic system to visual servo control in case of an uncooperative tumbling object. In contrast to the previously studied approaches that require a 3D CAD model of the object or its reconstruction, we propose a novel solution that also forgoes the need for special markers. For this purpose, we leverage a deep convolutional neural network technique to automatically estimate the axis of rotation vector of a tumbling object from its video and motion characteristics. Position-Based Visual Servoing algorithm can then use the extracted data for control. The effectiveness of the proposed framework is exhibited by implementing simulation in V-Rep on the Reachy Robotic arm.

Autonomous assembly of objects is an essential task in robotics and 3D computer vision. It has been studied extensively in robotics as a problem of motion planning, actuator control and obstacle avoidance. However, the task of developing a generalized framework for assembly robust to structural variants remains relatively unexplored. In this work, we tackle this problem using a recurrent graph learning framework considering inter-part relations and the progressive update of the part pose. Our network can learn more plausible predictions of shape structure by accounting for priorly assembled parts. Compared to the current state-of-the-art, our network yields up to 10% improvement in part accuracy and up to 15% improvement in connectivity accuracy on the PartNet [23] dataset. Moreover, our resulting latent space facilitates exciting applications such as shape recovery from the point-cloud components. We conduct extensive experiments to justify our design choices and demonstrate the effectiveness of the proposed framework.

Reflection symmetry is a very commonly occurring feature in both natural and man-made objects, which helps in understanding objects better and makes them visually pleasing. Detection of reflection symmetry is a fundamental problem in the field of computer vision and computer graphics which aids in understanding and representing reflective symmetric objects. In this work, we attempt the problem of detecting the 3D global reflection symmetry of a 3D object represented as a point cloud. The main challenge is to handle outliers, missing parts, and perturbations from the perfect reflection symmetry. We propose a descriptor-free approach, in which, we pose the problem of reflection symmetry detection as an optimization problem and provide a closed-form solution. We show that the proposed method achieves state-of-the-art performance on the standard dataset.

Continuous spread of novel coronavirus (COVID-19) and availability of limited resources force the severity-based allocation of resources. While it is essential to have a reliable severity assessment method, it is even more critical to have a prognosis model to estimate infection progress in individuals. An accurate estimate of infection progression would naturally help in optimized treatment and morbidity reduction. We aim at the prognosis of the COVID-19 infections including, ground-glass opacities, consolidation, and pleural effusion, from the longitudinal chest X-ray (CXR) images of the patient. For this purpose, we first propose a learning-based framework that predicts infection type from a given CXR image. This helps in finding low dimensional embeddings of CXR images, which we use in a recurrent learning framework to predict the type of infection for the subsequent days. We achieve a test AUC of 0.85 for infection type prediction and a test AUC of 0.88 for prognosis on the benchmark COVID-19 dataset.

With the rise of deep reinforcement learning (RL) methods, many complex robotic manipulation tasks are being solved. However, harnessing the full power of deep learning requires large datasets. Online RL does not suit itself readily into this paradigm due to costly and time-consuming agent-environment interaction. Therefore, many offline RL algorithms have recently been proposed to learn robotic tasks. But mainly, all such methods focus on a single-task or multitask learning, which requires retraining whenever we need to learn a new task. Continuously learning tasks without forgetting previous knowledge combined with the power of offline deep RL would allow us to scale the number of tasks by adding them one after another. This paper investigates the effectiveness of regularisation-based methods like synaptic intelligence for sequentially learning image-based robotic manipulation tasks in an offline-RL setup. We evaluate the performance of this combined framework against common challenges of sequential learning: catastrophic forgetting and forward knowledge transfer. We performed experiments with different task combinations to analyse the effect of task ordering. We also investigated the effect of the number of object configurations and the density of robot trajectories. We found that learning tasks sequentially helps in the retention of knowledge from previous tasks, thereby reducing the time required to learn a new task. Regularisation-based approaches for continuous learning, like the synaptic intelligence method, help mitigate catastrophic forgetting but have shown only limited transfer of knowledge from previous tasks.