Search Constraints

Start Over
Filtering by: Creator Cross, Tim Remove constraint Creator: Cross, Tim Committee Member Li, Fuxin Remove constraint Committee Member: Li, Fuxin Degree Name Doctor of Philosophy (Ph.D.) Remove constraint Degree Name: Doctor of Philosophy (Ph.D.)

Search Results

  • Action Segmentation with Limited Supervision

    Creator:
    Li, Jun
    Abstract:
    In this dissertation, we address action segmentation in videos under limited supervision. The goal of action segmentation is to predict an action class for each frame of a video. The limited supervision means ground truth labels of video frames are not available in training. We focus on three types of...
    Resource Type:
    Dissertation
    Full Text:
    for computing the cross-entropy loss and n-pair loss. . . . . . . . . . . . . . . . . 37 4.9 Our SCV

  • Pixel- and Frame-level Video Labeling using Spatial and Temporal Convolutional Networks

    Creator:
    Lei, Peng
    Abstract:
    This dissertation addresses the problem of video labeling at both the frame and pixel levels using deep learning. For pixel-level video labeling, we have studied two problems: i) Spatiotemporal video segmentation and ii) Boundary detection and boundary flow estimation. For the problem of spatiotemporal video segmentation, we have developed recurrent...
    Resource Type:
    Dissertation
    Full Text:
    are given from a 5-fold cross-validation. . . . . . . . . 17 2.2 Pixel accuracy on Cambridge-driving

  • Semantic embodied navigation : developing agents that navigate from language and vision

    Creator:
    Krantz, Jacob
    Abstract:
    Autonomous robotic agents are on their way to becoming in-home personal assistants, construction assistants, and warehouse workers. The degree of autonomy of such systems is reflected by the manner in which we specify goals to them; the abstraction of low-level commands to high-level goals goes hand-in-hand with increased autonomy. In...
    Resource Type:
    Dissertation
    Full Text:
    -to-Sequence Baseline . . . . . . . . . . . . . . . . . . . . 24 2.4.2 Cross-Modal Attention Model

  • Ideal Thumbnail-Preserving Encryption for Balancing Image Privacy and Usability

    Creator:
    Tajik, Kimia
    Abstract:
    In this dissertation, we propose Ideal Thumbnail-Preserving Encryption (Ideal TPE), as a special case of format-preserving encryption, to balance image privacy and usability concerns in a cloud environment. We first introduce a concrete construction for Ideal TPE, that provably leaks nothing about the plaintext (unencrypted) image beyond its thumbnail. We...
    Resource Type:
    Dissertation
    Full Text:
    privacy threat to TPE. Cross-Resolution Face Recognition: Another category of solutions for low

  • Robust and Efficient Classification of Videos in the Wild

    Creator:
    Mahasseni, Behrooz
    Abstract:
    Recognizing human actions in videos is a long-standing problem in computer vision with a wide range of applications including video surveillance, content retrieval, and sports analysis. This thesis focuses on addressing efficiency and robustness of video classification in unconstrained real-world settings. The thesis work can be broadly divided into four...
    Resource Type:
    Dissertation
    Full Text:
    BMX, cyclo- cross and road bicycle racing are similar actions which all contain humans riding a bike

  • Investigating Latent State and Uncertainty Representations in Reinforcement Learning

    Creator:
    Koul, Anurag
    Abstract:
    Learning latent space representations of high-dimensional world states has been at the core of recent rapid growth in reinforcement learning(RL). At the same time, RL algo- rithms have suffered from ignored uncertainties in the predicted estimates of model-free or model-based methods. In our work, we investigate both of these aspects...
    Resource Type:
    Dissertation

  • Fine-Grained Object Recognition Under Limited Training Data

    Creator:
    Lam, Michael Q.
    Abstract:
    This dissertation addresses object recognition in challenging settings, where distinct object classes are visually very similar (e.g., species of birds and insects) and/or access to training examples of object classes is limited (e.g., due to the associated high costs of data annotation). In this dissertation, we present a variety of...
    Resource Type:
    Dissertation
    Full Text:
    five-fold cross vali- dation experiment setup in prior work [43]. For the Make3D dataset, we employed

  • Semantic image segmentation using domain constraints

    Creator:
    Roy, Anirban
    Abstract:
    This dissertation addresses the problem of semantic labeling of image pixels. In the course of our work, we considered different types of semantic labels, including object classes (e.g., car, person), 3D depth values (in the range 0 to 80 meters), and affordance classes (e.g., walkable, sittable). Semantic pixel labeling is...
    Resource Type:
    Dissertation

  • Explanations and Processes to Enable Humans to Assess AI with Respect to Manipulable Properties

    Creator:
    Dodge, Jonathan E.
    Abstract:
    Assessing AI systems is difficult. Humans rely on AI systems in increasing ways, both visible and invisible, meaning a variety of stakeholders need a variety of assessment tools (e.g., a professional auditor, a developer, and an end user all have different needs). We posit that it is possible to provide...
    Resource Type:
    Dissertation