Abstract

We present DRACO, a method for Dense Reconstruction And Canonicalization of Object shape from one or more RGB images. Canonical shape reconstruction, estimating 3D object shape in a coordinate space canonicalized for scale, rotation, and translation parameters, is an emerging paradigm that holds promise for a multitude of robotic applications. Prior approaches either rely on painstakingly gathered dense 3D supervision, or produce only sparse canonical representations, limiting real-world applicability. DRACO performs dense canonicalization using only weak supervision in the form of camera poses and semantic keypoints at train time. During inference, DRACO predicts dense object-centric depth maps in a canonical coordinate-space, solely using one or more RGB images of an object. Extensive experiments on canonical shape reconstruction and pose estimation show that DRACO is competitive or superior to fully-supervised methods.

Abstract

Video Prediction is an interesting and challenging task of predicting future frames from a given set context frames that belong to a video sequence. Video prediction models have found prospective applications in Maneuver Planning, Health care, Autonomous Navigation and Simulation. One of the major challenges in future frame generation is due to the high dimensional nature of visual data. In this work, we propose Mutual Information Predictive Auto-Encoder (MIPAE) framework, that reduces the task of predicting high dimensional video frames by factorising video representations into content and low dimensional pose latent variables that are easy to predict. A standard LSTM network is used to predict these low dimensional pose representations. Content and the predicted pose representations are decoded to generate future frames. Our approach leverages the temporal structure of the latent generative factors of a video and a novel mutual information loss to learn disentangled video representations. We also propose a metric based on mutual information gap (MIG) to quantitatively access the effectiveness of disentanglement on DSprites and MPI3D-real datasets. MIG scores corroborate with the visual superiority of frames predicted by MIPAE. We also compare our method quantitatively on evaluation metrics LPIPS, SSIM and PSNR.

Abstract

In this paper, we present BirdSLAM, a novel simultaneous localization and mapping (SLAM) system for the challenging scenario of autonomous driving platforms equipped with only a monocular camera. BirdSLAM tackles challenges faced by other monocular SLAM systems (such as scale ambiguity in monocular reconstruction, dynamic object localization, and uncertainty in feature representation) by using an orthographic (bird's-eye) view as the configuration space in which localization and mapping are performed. By assuming only the height of the ego-camera above the ground, BirdSLAM leverages single-view metrology cues to accurately localize the ego-vehicle and all other traffic participants in bird's-eye view. We demonstrate that our system outperforms prior work that uses strictly greater information, and highlight the relevance of each design decision via an ablation analysis.

Abstract

Modern NLP defines the task of style transfer as modifying the style of a given sentence without appreciably changing its semantics, which implies that the outputs of style transfer systems should be paraphrases of their inputs. However, many existing systems purportedly designed for style transfer inherently warp the input's meaning through attribute transfer, which changes semantic properties such as sentiment. In this paper, we reformulate unsupervised style transfer as a paraphrase generation problem, and present a simple methodology based on fine-tuning pretrained language models on automatically generated paraphrase data. Despite its simplicity, our method significantly outperforms state-of-the-art style transfer systems on both human and automatic evaluations. We also survey 23 style transfer papers and discover that existing automatic metrics can be easily gamed and propose fixed variants. Finally, we pivot to a more real-world style transfer setting by collecting a large dataset of 15M sentences in 11 diverse styles, which we use for an in-depth analysis of our system.

Abstract

Manual identification of ECG heart-beat classes by cardiologists is time consuming and cumbersome. These professionals rely on computer based methods for determination of these heart-disease types. In this work, existing literature is organized into a proposed taxonomy based on dichotomies involving full time series-based versus feature-based, AAMI versus Non-AAMI, and inter-patient versus intra-patient based distinctions. The basic contributions of this work are systematic review of literature on heart-beat abnormality detection, identifying research gaps and the research issues unmet sofar in the literature to propose novel approaches for addressing these gaps.

Abstract

Executing a robust person-following behaviour is a crucial task for service robots. This requires real-time tracking of the target person and maintaining it in its field of view (FOV) as long as possible while also keeping a safe distance without collision. Such a robot has to autonomously traverse long corridors, bend around corners, curve around an obstacle and join the person back, or even wait them out. Thus, in an unknown urban setting, with other agents and service robots, dynamic target tracking and obstacle avoidance becomes a challenging task for autonomous non-holonomic robots. Model Predictive Control (MPC) has the ability to incorporate future predictions. vehicle kinematics, and non-linearity of constraints while still being reasonably fast for slow and medium walking speeds of a human being. Thus, by solving a single non-linear MPC we can track the unknown dynamic trajectory of a person in real-time, and avoid undesirable proximity. In this paper, a constrained MPC framework is designed that meets the above conditions gracefully. The person-following robot (PFR) is able to track a moving target in a simulated environment. With an adequate upper margin of 20Hz, the MPC sends velocity commands for a collision-free trajectory among an increasing number of obstacles in its perceived periphery. In an incremental fashion, it is also able to adapt to varying walking behaviour, that is, types of trajectories and walking speed.

Abstract

Significant advances have been made recently in Visual Place Recognition (VPR), feature correspondence, and localization due to the proliferation of deep-learning-based methods. However, existing approaches tend to address, partially or fully, only one of two key challenges: viewpoint change and perceptual aliasing. In this paper, we present novel research that simultaneously addresses both challenges by combining deep-learned features with geometric transformations based on reasonable domain assumptions about navigation on a ground-plane, whilst also removing the requirement for specialized hardware setup (eg lighting, downwards facing cameras). In particular, our integration of VPR with SLAM by leveraging the robustness of deep-learned features and our homography-based extreme viewpoint invariance significantly boosts the performance of VPR, feature correspondence, and pose graph submodules of the SLAM pipeline. For the first time, we demonstrate a localization system capable of state-of-the-art performance despite perceptual aliasing and extreme 180-degree-rotated viewpoint change in a range of real-world and simulated experiments. Our system is able to achieve early loop closures that prevent significant drifts in SLAM trajectories. We also compare extensively several deep architectures for VPR and descriptor matching. We also show that superior place recognition and descriptor matching across opposite views results in a similar performance gain in back-end pose graph optimization.

Abstract

Music, an integral part of our lives, which is not only a source of entertainment but plays an important role in mental well-being by impacting moods, emotions and other affective states. Music preferences and listening strategies have been shown to be associated with the psychological well-being of listeners including internalized symptomatology and depression. However, till date no studies exist that examine time-varying music consumption, in terms of acoustic content, and its association with users' well-being. In the current study, we aim at unearthing static and dynamic patterns prevalent in active listening behavior of individuals which may be used as indicators of risk for depression. Mental well-being scores and listening histories of 541 this http URL users were examined. Static and dynamic acoustic and emotion-related features were extracted from each user's listening history and correlated with their mental well-being scores. Results revealed that individuals with greater depression risk resort to higher dependency on music with greater repetitiveness in their listening activity. Furthermore, the affinity of depressed individuals towards music that can be perceived as sad was found to be resistant to change over time. This study has large implications for future work in the area of assessing mental illness risk by exploiting digital footprints of users via online music streaming platforms.

Abstract

In this paper, we present a simple baseline for visual grounding for autonomous driving which outperforms the state of the art methods, while retaining minimal design choices. Our framework minimizes the cross-entropy loss over the cosine distance between multiple image ROI features with a text embedding (representing the given sentence/phrase). We use pre-trained networks for obtaining the initial embeddings and learn a transformation layer on top of the text embedding. We perform experiments on the Talk2Car dataset and achieve 68.7% AP50 accuracy, improving upon the previous state of the art by 8.6%. Our investigation suggests reconsideration towards more approaches employing sophisticated attention mechanisms or multi-stage reasoning or complex metric learning loss functions by showing promise in simpler alternatives.

Abstract

Recent neuroimaging evidence suggest that there exists a unique individual-specific functional connectivity (FC) pattern consistent across tasks. The objective of our study is to utilize FC patterns to identify an individual using a supervised machine learning approach. To this end, we use two previously published data sets that comprises resting-state and task-based fMRI responses. We use static FC measures as input to a linear classifier to evaluate its performance. We additionally extend this analysis to capture dynamic FC using two approaches: the common sliding window approach and the more recent phase synchrony-based measure. We found that the classification models using dynamic FC patterns as input outperform their static analysis counterpart by a significant margin for both data sets. Furthermore, sliding window-based analysis proved to capture more individual-specific brain connectivity patterns …