MSR Thesis Defense

Title: Tactile Sensing applied to deformable object manipulation Abstract: The application of robotic manipulation of deformable materials, such as cloth, spans various sectors including fabric manufacturing and domestic laundry management. Historically, most methodologies have employed vision-based sensors as the proprioceptive input to robot policies. However, this study aims to explore an alternate route by leveraging [...]

Title: Distilling View-conditioned Diffusion for 3D Reconstruction Abstract: We propose a 3D neural mode-seeking formulation that combines probabilistic generation of unseen regions with faithful reprojection of seen regions in a consistent 3D representation. Feature reprojection methods (NerFormer, PixelNeRF) are 3D consistent, but fail to hallucinate unseen regions. Image generation methods (ViewFormer) generate plausible hallucinations, but generated [...]

Title: Scaling up Camera Calibration and Amodal 3D Object Reconstruction for Smart Cities Abstract: Smart cities integrate thousands of outdoor cameras to enhance urban infrastructure, but their automated analysis potential remains untapped due to various challenges. Firstly, the lack of accurate camera calibration information, such as its intrinsics parameters and external orientation, restricts the measurement [...]

Title: Surface Ripples: Analyzing Transient Vibrations on Object's Surfaces Abstract: The subtle vibrations on an object's surface contain information about its physical properties and interaction with the environment.  Prior works imaged surface vibration to recover the object's material properties via modal analysis, which discards the transient vibrations propagating immediately after the object is disturbed. In this [...]

Title: Learned Two-Plane Perspective Prior based Image Resampling for Efficient Object Detection Abstract:    Real-time efficient perception is critical for autonomous navigation and city scale sensing. Orthogonal to architectural improvements, streaming perception approaches have exploited adaptive sampling improving real-time detection performance. In this work, we propose a learnable geometry-guided prior that incorporates rough geometry of the [...]

Title: Using Drones and Remote Sensing to Understand Forests with Limited Labeled Data Abstract: Drones and remote sensing can provide observations of forests at scale, but this raw data needs to be interpreted to further scientific understanding and inform effective management decisions. This thesis studies two problems under the realistic constraint of limited domain-specific training [...]

Title: Lights, Camera, Render: Neural Fields for Structured Lighting Abstract: 3D scene reconstruction from 2D image supervision alone is an under-constrained problem. Recent neural rendering frameworks have made great strides in learning 3D scene representations to enable novel view synthesis, but they struggle to reconstruct geometry of low-texture regions or from sparse views. The prevalence of active [...]

Title: Learning to See in the Dark and Beyond Abstract: Robotic Perception in diverse domains such as low-light scenarios remains a challenge, even upon the employment of new sensing modalities like thermal imaging and specialized night-vision sensors. This is largely due to the high difficulty in obtaining labeled data for certain tasks. In this work, [...]

Title: Self-Supervised Costmap Learning for Off-Road Vehicle Traversability Abstract: Estimating terrain traversability in off-road environments requires reasoning about complex interaction dynamics between the robot and these terrains. However, it is challenging to build an accurate physics model, or create informative labels to learn a model in a supervised manner, for these interactions. We propose a method [...]