Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 23 for “"Scene flow"”.

  1. Scene flow for autonomous navigation

    … perception approaches that parse an entire scene to make driving decision, a direct perception approach that estimates the affordance of driving that maps an input image to small number of key perception indicators that directly relate to the affordance of road/traffic state for driving. …

    gatech Repository record for Scene flow for autonomous navigation (opens in a new tab)

  2. RGB-D Scene Flow via Grouping Rigid Motions

    … environments is analyzing the motions within the scene, otherwise known as scene flow. Scene flow estimates the 3D velocity of each imaged point captured by a camera. The 3D information of the scene can be acquired by RGB-D cameras, which produce both colour and depth images and have been proven …

    waterloo Repository record for RGB-D Scene Flow via Grouping Rigid Motions (opens in a new tab)

  3. Toward Scalable, Flexible Scene Flow for Point Clouds

    Scene flow estimation is the task of describing 3D motion between temporally successive observations. This dissertation aims to build the foundation for building scene flow estimators with two important properties: they are scalable, i.e. they improve with access to more data and computation, and …

    penn Repository record for Toward Scalable, Flexible Scene Flow for Point Clouds (opens in a new tab)

  4. Empower dynamic scene understanding through scene flow estimation and object segmentation

    Understanding dynamic 3D scenes—critical for applications like autonomous navigation and mixed reality—requires pars- ing both motion (scene flow) and object interactions (segmen- tation). Scene flow captures 3D motion fields, while segmen- tation isolates objects, enabling systems to interpret …

    bournemouth Download PDF for Empower dynamic scene understanding through scene flow estimation and object segmentation (opens in a new tab)

  5. Understanding Motion and Geometry of Point Cloud in Dynamic Environment

    … shapes, rigid poses, and the motion of dynamic scenes, etc.. Understanding dynamic environments through learning the geometry and motion is critical for robotic applications. However, the inherent challenges of 3D data—such as its discrete, unordered structure, variable densities (e.g., sparse …

    adelaide Repository record for Understanding Motion and Geometry of Point Cloud in Dynamic Environment (opens in a new tab)

  6. Motion Learning for Dynamic Scene Understanding

    … Motion plays a key role in analyzing dynamic scenes and has been one of the fundamental research topics in computer vision. It has wide applications in many fields, including video analysis, socially-aware robotics, autonomous driving, etc. In this dissertation, we study motion from two …

    jhu Repository record for Motion Learning for Dynamic Scene Understanding (opens in a new tab)

  7. VISUAL DENSE THREE-DIMENSIONAL MOTION ESTIMATION IN THE WILD

    … continuous optimization approach that solves the scene motions as fixed-a-priori planar segments. 2) Present a learning-based approach that recovers the dense scene flow from egocentric motion and optical flow, decomposed by a novel data-driven rigidity prediction. 3) Present a modern synthesis of …

    gatech Repository record for VISUAL DENSE THREE-DIMENSIONAL MOTION ESTIMATION IN THE WILD (opens in a new tab)

  8. Machine vision analysis of vehicle interiors and surrounding road users in intelligent transportation

    … VGG16 network first classifies rear-seat scenes as clean, trash or valuables with 0.91 accuracy. The architecture is then extended with an EfficientDet-based detector which yielded prediction accuracies of 0.89-0.93 across broader variability (such as vehicle models and item types). The …

    aalto Repository record for Machine vision analysis of vehicle interiors and surrounding road users in intelligent transportation (opens in a new tab)

  9. Robust Dynamic Vision from a Moving Platform

    … of the 3D distance and movement from a scene. This is substantially more difficult when the scenes are captured from a "moving platform" as there are also independent moving objects along with the movement from the scene. This thesis provides a solution to the issue of robustness by …

    auckland-ms Repository record for Robust Dynamic Vision from a Moving Platform (opens in a new tab)

  10. Cooperative Path Planning for Autonomous Ground Vehicles using 3D Sensor in Cluttered Environment

    … An approach which combines the differential scene flow technique and the gradient vector field (GVF) for estimating the kinematic behaviours of the obstacles using the range intensity value of the PMD camera is proposed. To ease the complexity and spatial sparseness in single AGV system in …

    unsw Repository record for Cooperative Path Planning for Autonomous Ground Vehicles using 3D Sensor in Cluttered Environment (opens in a new tab)

  11. Convolutional networks to relate images

    … end-to-end learning approach to compute optical flow and extend it to disparity and scene flow.<br>All of this is done by using synthetically generated training data which we rendered using a custom version of the Blender rendering suite.

    freiburg-diss Repository record for Convolutional networks to relate images (opens in a new tab)

  12. Gradient Consistency: A New Take on Variational Optical Flow and Disparity Estimation

    Optical flow and disparity estimation aim to determine a displacement field between images of the same scene. Knowledge-driven approaches to these tasks revolve around brightness constancy assumptions, i.e. objects when displaced do not change in appearance. Variational approaches to optical flow …

    unsw Repository record for Gradient Consistency: A New Take on Variational Optical Flow and Disparity Estimation (opens in a new tab)

  13. 3-D Object Segmentation from Motion Cues

    … a 2-D view or full 3-D, of objects present in a scene. In this work we make use of motion caused both by the robot and by other agents to perform segmentation. Similarly, we can use motion for segmentation whether the actual movement was observed by a robot or not. Our motion estimation …

    washington Repository record for 3-D Object Segmentation from Motion Cues (opens in a new tab)

  14. Robust, optimal, safe, and stable visual-inertial based navigation of autonomous vehicle with neural network in feedback loop

    … segmentation with object detection and optical flow to classify image regions as static or dynamic, based on semantic labels and pixel movements. Optical flow gradients are leveraged to mitigate the influence of camera motion, and a consistency check based on scene flow further refines the …

    iastate Repository record for Robust, optimal, safe, and stable visual-inertial based navigation of autonomous vehicle with neural network in feedback loop (opens in a new tab)

  15. Control of Low Speed Flight in Cluttered Environments Using Image Motion and Ranging on a Micro-Air Vehicle

    … Algorithm (I2A) is used to calculate optical flow from the RGB-D camera intensity image sequences and corresponding depth information. A novel ego-motion recovery method is proposed by combining the range data with the optical flow field Image Jacobian. For our algorithm, assumptions about the …

    unsw Repository record for Control of Low Speed Flight in Cluttered Environments Using Image Motion and Ranging on a Micro-Air Vehicle (opens in a new tab)

  16. Robust 2D and 3D registration with deep neural networks

    … minima. We introduce a neural network-based scene flow to initialize the optimization, providing a more efficient and robust solution. Additionally, we present a novel surface normal estimation technique that aids both rigid and non-rigid registration. Unlike conventional methods that use a …

    oxford Repository record for Robust 2D and 3D registration with deep neural networks (opens in a new tab)

  17. Detection of Moving Objects by Spatio-Temporal Motion Analysis

    … obtain the image measurements, various optical flow and stereo algorithms are evaluated. In particular, a novel dense stereo algorithm is presented, that gives excellent precision results and runs at real-time. In addition, two novel scene flow algorithms are introduced, that measure the optical …

    cau-kiel Repository record for Detection of Moving Objects by Spatio-Temporal Motion Analysis (opens in a new tab)

  18. Deep representation learning for dynamic point cloud sequences

    … point cloud processing methods assume static scenes. This fundamental limitation restricts the deployment of 3D perception systems in applications requiring temporal understanding, from autonomous navigation to human-robot interaction. This thesis presents three techniques for dynamic point …

    oxford Repository record for Deep representation learning for dynamic point cloud sequences (opens in a new tab)

  19. Learning Depth From Images

    … in computer vision which aims to restore the 3D scene from 2D images and identify important geometric knowledge of the scene. Its performance has been significantly improved by convolutional neural networks in recent years, which surpass the traditional methods by a large margin. However, the …

    south-carolina Repository record for Learning Depth From Images (opens in a new tab)

  20. Real-time Visual Flow Algorithms for Robotic Applications

    … addresses the problem of estimating dense visual flow information in real time. The contributions of this work are threefold. First, it introduces a new filtering algorithm for the estimation of dense optical flow at frame rates as fast as 800 Hz for 640x480 image resolution. The algorithm follows …

    anu Repository record for Real-time Visual Flow Algorithms for Robotic Applications (opens in a new tab)

Page 1 of 2