Jump to Content

Thomas Funkhouser

Research Areas

Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, desc
  • Year
  • Year, desc
    Panoptic Neural Fields: A Semantic Object-Aware Neural Scene Representation
    Kyle Genova
    Xiaoqi Yin
    Leonidas Guibas
    Frank Dellaert
    Conference on Computer Vision and Pattern Recognition (2022)
    Preview abstract We present Panoptic Neural Fields (PNF), an object-aware neural scene representation that decomposes a scene into a set of objects (things) and background (stuff). Each object is represented by an oriented 3D bounding box and a multi-layer perceptron (MLP) that takes position, direction, and time and outputs density and radiance. The background stuff is represented by a similar MLP that additionally outputs semantic labels. Each object MLPs are instance-specific and thus can be smaller and faster than previous object-aware approaches, while still leveraging category-specific priors incorporated via meta-learned initialization. Our model builds a panoptic radiance field representation of any scene from just color images. We use off-the-shelf algorithms to predict camera poses, object tracks, and 2D image semantic segmentations. Then we jointly optimize the MLP weights and bounding box parameters using analysis-by-synthesis with self-supervision from color images and pseudo-supervision from predicted semantic segmentations. During experiments with real-world dynamic scenes, we find that our model can be used effectively for several tasks like novel view synthesis, 2D panoptic segmentation, 3D scene editing, and multiview depth prediction. View details
    Learning Pneumatic Non-Prehensile Manipulation with a Mobile Blower
    Jimmy Wu
    Xingyuan Sun
    Shuran Song
    Szymon Rusinkiewicz
    IEEE Robotics and Automation Letters (RA-L) (2022)
    Preview abstract We investigate pneumatic non-prehensile manipulation (i.e., blowing) as a means of efficiently moving scattered objects into a target receptacle. Due to the chaotic nature of aerodynamic forces, a blowing controller must (i) continually adapt to unexpected changes from its actions, (ii) maintain fine-grained control, since the slightest misstep can result in large unintended consequences (e.g., scatter objects already in a pile), and (iii) infer long-range plans (e.g., move the robot to strategic blowing locations). We tackle these challenges in the context of deep reinforcement learning, introducing a multi-frequency version of the spatial action maps framework. This allows for efficient learning of vision-based policies that effectively combine high-level planning and low-level closed-loop control for dynamic mobile manipulation. Experiments show that our system learns efficient behaviors for the task, demonstrating in particular that blowing achieves better downstream performance than pushing, and that our policies improve performance over baselines. Moreover, we show that our system naturally encourages emergent specialization between the different subpolicies spanning low-level fine-grained control and high-level planning. On a real mobile robot equipped with a miniature air blower, we show that our simulation-trained policies transfer well to a real environment and can generalize to novel objects. View details
    Preview abstract A classical problem in computer vision is to infer a 3D scene representation from few images that can be used to render novel views at interactive rates. Previous work focuses on reconstructing pre-defined 3D representations, e.g. textured meshes, or implicit representations, e.g. radiance fields, and often requires input images with precise camera poses and long processing times for each novel scene. In this work, we propose the Scene Representation Transformer (SRT), a method which processes posed or unposed RGB images of a new area, infers a "set-latent scene representation", and synthesises novel views, all in a single feed-forward pass. To calculate the scene representation, we propose a generalization of the Vision Transformer to sets of images, enabling global information integration, and hence 3D reasoning. An efficient decoder transformer parameterizes the light field by attending into the scene representation to render novel views. Learning is supervised end-to-end by minimizing a novel-view reconstruction error. We show that this method outperforms recent baselines in terms of PSNR and speed on synthetic datasets, including a new dataset created for the paper. Further, we demonstrate that SRT scales to support interactive visualization and semantic segmentation of real-world outdoor environments using Street View imagery. View details
    Spatial Intention Maps for Multi-Agent Mobile Manipulation
    Jimmy Wu
    Xingyuan Sun
    Shuran Song
    Szymon Rusinkiewicz
    IEEE International Conference on Robotics and Automation (ICRA) (2021)
    Preview abstract The ability to communicate intention enables decentralized multi-agent robots to collaborate while performing physical tasks. In this work, we present spatial intention maps, a new intention representation for multi-agent vision-based deep reinforcement learning that improves coordination between decentralized mobile manipulators. In this representation, each agent's intention is provided to other agents, and rendered into an overhead 2D map aligned with visual observations. This synergizes with the recently proposed spatial action maps framework, in which state and action representations are spatially aligned, providing inductive biases that encourage emergent cooperative behaviors requiring spatial coordination, such as passing objects to each other or avoiding collisions. Experiments across a variety of multi-agent environments, including heterogeneous robot teams with different abilities (lifting, pushing, or throwing), show that incorporating spatial intention maps improves performance for different mobile manipulation tasks while significantly enhancing cooperative behaviors. View details
    Multiresolution Deep Implicit Functions for 3D Shape Representation
    Zhang Chen
    Kyle Genova
    Sofien Bouaziz
    Christian Haene
    Cem Keskin
    Danhang "Danny" Tang
    ICCV (2021)
    Preview abstract We introduce Multiresolution Deep Implicit Functions (MDIF), a hierarchical representation that can recover fine details, while being able to perform more global operations such as shape completion. Our model represents a complex 3D shape with a hierarchy of latent grids, which can be decoded into different resolutions. Training is performed in an encoder-decoder manner, while the decoder-only optimization is supported during inference, hence can better generalize to novel objects, especially when performing shape completion. To the best of our knowledge, MDIF is the first model that can at the same time (1) reconstruct local detail; (2) perform decoder-only inference; (3) fulfill shape reconstruction and completion. We demonstrate superior performance against prior arts in our experiments. View details
    IBRNet: Learning Multi-View Image-Based Rendering
    Kyle Genova
    Pratul Srinivasan
    Qianqian Wang
    Ricardo Martin-Brualla
    Zhicheng Wang
    Conference on Computer Vision and Pattern Recognition (CVPR), IEEE (2021) (to appear)
    Preview abstract We present a method that synthesizes novel views of complex scenes by interpolating a sparse set of nearby views.The core of our method is a multilayer perceptron (MLP)that generates RGBA at each 5D coordinate from multi-view image features. Unlike neural scene representation work that optimizes per-scene functions for rendering, we learn a generic view interpolation function that naturally generalizes to novel scene types and camera setups. Compared to previous generic image-based rendering (IBR) methods like Multiple-plane images (MPIs) that use discrete volume representations, our method instead produces RGBAs at continuous 5D locations (3D spatial locations and 2D viewing directions), enabling high-resolution imagery rendering.Our rendering pipeline is fully differentiable, and the only input required to train our method are multi-view posed images. Experiments show that our method outperforms previous IBR methods, and achieves state-of-the-art performance when fine tuned on each test scene. View details
    Preview abstract Detecting objects in 3D LiDAR data is a core technology for autonomous driving and other robotics applications. Although LiDAR data is acquired over time, most of the 3D object detection algorithms propose object bounding boxes independently for each frame and neglect the useful information available in the temporal domain. To address this problem, in this paper we propose a sparse LSTM-based multi-frame 3d object detection algorithm. We use a U-Net style 3D sparse convolution network to extract features for each frame's LiDAR point-cloud. These features are fed to the LSTM module together with the hidden and memory features from last frame to predict the 3d objects in the current frame as well as hidden and memory features that are passed to the next frame. Experiments on the Waymo Open Dataset show that our algorithm outperforms the traditional frame by frame approach by 7.5% mAP@0.7 and other multi-frame approaches by 1.2% while using less memory and computation per frame. To the best of our knowledge, this is the first work to use an LSTM for 3D object detection in sparse point clouds. View details
    Preview abstract Semantic segmentation of 3D meshes is an important problem for 3D scene understanding. In this paper we revisit the classic multiview representation of 3D meshes and study several techniques that make them effective for 3D semantic segmentation of meshes. Given a 3D mesh reconstructed from RGBD sensors, our method effectively chooses different virtual views of the 3D mesh and renders multiple 2D channels for training an effective 2D semantic segmentation model. Features from multiple per view predictions are finally fused on 3D mesh vertices to predict mesh semantic segmentation labels. Using the large scale indoor 3D semantic segmentation benchmark of ScanNet, we show that our virtual views enable more effective training of 2D semantic segmentation networks than previous multiview approaches. When the 2D per pixel predictions are aggregated on 3D surfaces, our virtual multiview fusion method is able to achieve significantly better 3D semantic segmentation results compared to all prior multiview approaches and competitive with recent 3D convolution approaches. View details
    Spatial Action Maps for Mobile Manipulation
    Jimmy Wu
    Xingyuan Sun
    Shuran Song
    Szymon Rusinkiewicz
    Robotics: Science and Systems (RSS) (2020)
    Preview abstract Typical end-to-end formulations for learning robotic navigation involve predicting a small set of steering command actions (e.g., step forward, turn left, turn right, etc.) from images of the current state (e.g., a bird's-eye view of a SLAM reconstruction). Instead, we show that it can be advantageous to learn with dense action representations defined in the same domain as the state. In this work, we present "spatial action maps," in which the set of possible actions is represented by a pixel map (aligned with the input image of the current state), where each pixel represents a local navigational endpoint at the corresponding scene location. Using ConvNets to infer spatial action maps from state images, action predictions are thereby spatially anchored on local visual features in the scene, enabling significantly faster learning of complex behaviors for mobile manipulation tasks with reinforcement learning. In our experiments, we task a robot with pushing objects to a goal location, and find that policies learned with spatial action maps achieve much better performance than traditional alternatives. View details
    Preview abstract We propose DOPS, a fast single-stage 3D object detection method for LIDAR data. Previous methods often make domain-specific design decisions, for example projecting points into a bird-eye view image in autonomous driving scenarios. In contrast, we propose a general-purpose method that works on both indoor and outdoor scenes. The core novelty of our method is a fast, single-pass architecture that both detects objects in 3D and estimates their shapes. 3D bounding box parameters are estimated in one pass for every point, aggregated through graph convolutions, and fed into a branch of the network that predicts latent codes representing the shape of each detected object. The latent shape space and shape decoder are learned on a synthetic dataset and then used as supervision for the end-toend training of the 3D object detection pipeline. Thus our model is able to extract shapes without access to groundtruth shape information in the target dataset. During experiments, we find that our proposed method achieves stateof-the-art results by ∼5% on object detection in ScanNet scenes, and it gets top results by 3.4% in the Waymo Open Dataset, while reproducing the shapes of detected cars. View details
    Preview abstract Intelligent manipulation benefits from the capacity to flexibly control an end-effector with high degrees of freedom (DoF) and dynamically react to the environment. However, due to the challenges of collecting effective training data and learning efficiently, most grasping algorithms today are limited to top-down movements and open-loop execution. In this work, we propose a new low-cost hardware interface for collecting grasping demonstrations by people in diverse environments. Leveraging this data, we show that it is possible to train a robust end-to-end 6DoF closed-loop grasping model with reinforcement learning that transfers to real robots. A key aspect of our grasping model is that it uses “action-view” based rendering to simulate future states with respect to different possible actions. By evaluating these states using a learned value function (Q-function), our method is able to better select corresponding actions that maximize total rewards (i.e., grasping success). Our final grasping system is able to achieve reliable 6DoF closed-loop grasping of novel objects across various scene configurations, as well as dynamic scenes with moving objects. View details
    Preview abstract We present a simple and flexible object detection framework optimized for autonomous driving. Building on the observation that point clouds in this application are extremely sparse, we propose a practical pillar-based approach to fix the imbalance issue caused by anchors. In particular, our algorithm incorporates a cylindrical projection into multi-view feature learning, predicts bounding box parameters per pillar rather than per point or per anchor, and includes an aligned pillar-to-point projection module to improve the final prediction. Our anchor-free approach avoids hyperparameter search associated with past methods, simplifying 3D object detection while significantly improving upon state-of-the-art. View details
    Preview abstract The goal of this project is to learn a 3D shape representation that enables accurate surface reconstruction, compact storage, efficient computation, consistency for similar shapes, generalization across diverse shape categories, and inference from depth camera observations. Towards this end, we introduce Local Deep Implicit Functions (LDIF), a 3D shape representation that decomposes space into a structured set of learned implicit functions. We provide networks that infer the space decomposition and local deep implicit functions from a 3D mesh or posed depth image. During experiments, we find that it provides 10.3 points higher surface reconstruction accuracy (F-Score) than the state-of-the-art (OccNet), while requiring fewer than 1 percent of the network parameters. Experiments on posed depth image completion and generalization to unseen classes show 15.8 and 17.8 point improvements over the state-of-the-art, while producing a structured 3D representation for each input with consistency across diverse shape collections. View details
    Preview abstract Throwing is a means to increase the capabilities of a manipulator by exploiting dynamics, a form of dynamic extrinsic dexterity. In the case of pick-and-place for example, throwing can enable a robot arm to rapidly place objects into selected boxes outside its maximum kinematic range — improving its physical reachability and picking speed. However, precisely throwing arbitrary objects in unstructured settings presents many challenges: from acquiring reliable pre-throw conditions (e.g. grasp of the object) to handling varying object-centric properties (e.g. mass distribution, friction, shape) and dynamics (e.g. aerodynamics). In this work, we propose an end-to-end formulation that jointly learns to infer control parameters for grasping and throwing motion primitives from visual observations (images of arbitrary objects in a bin) through trial and error. Within this formulation, we investigate the synergies between grasping and throwing (i.e., learning grasps that enable more accurate throws) and between simulation and deep learning (i.e. using deep networks to predict residuals on top of control parameters predicted by a physics simulator). The resulting system, TossingBot, is able to grasp and successfully throw arbitrary objects into boxes located outside its maximum reach range at 500+ mean picks per hour (600+ grasps per hour with 85% throwing accuracy); and generalizes to new objects and target locations. View details
    Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learning
    Shuran Song
    Stefan Welker
    Alberto Rodriguez
    IEEE International Conference on Intelligent Robots and Systems (IROS) (2018)
    Preview abstract Skilled robotic manipulation benefits from complex synergies between non-prehensile (e.g. pushing) and prehensile (e.g. grasping) actions: pushing can help rearrange cluttered objects to make space for arms and fingers; likewise, grasping can help displace objects to make pushing movements more precise and collision-free. In this work, we demonstrate that it is possible to discover and learn these synergies from scratch by combining visual affordance-based manipulation with model-free deep reinforcement learning. Our method is sample efficient and generalizes to novel objects and scenarios. View details
    ActiveStereoNet: Unsupervised End-to-End Learning for Active Stereo Systems
    Yinda Zhang
    Sameh Khamis
    Christoph Rhemann
    Julien Valentin
    Vladimir Tankovich
    Michael Schoenberg
    Shahram Izadi
    European Conference on Computer Vision (2018)
    Preview abstract In this paper we present ActiveStereoNet, the first deep learning solution for active stereo systems. Due to the lack of ground truth, our method is fully self-supervised, yet it produces precise depth with a subpixel precision of 1/30th of a pixel; it does not suffer from the common over-smoothing issues; it preserves the edges; and it explicitly handles occlusions. We introduce a novel reconstruction loss that is more robust to noise and texture-less patches, and is invariant to illumination changes. The proposed loss is optimized using a window-based cost aggregation with an adaptive support weight scheme. This cost aggregation is edge-preserving and smooths the loss function, which is key to allow the network to reach compelling results. Finally we show how the task of predicting invalid regions, such as occlusions, can be trained end-to-end without ground-truth. This component is crucial to reduce blur and particularly improves predictions along depth discontinuities. Extensive quantitatively and qualitatively evaluations on real and synthetic data demonstrate state of the art results in many challenging scenes. View details
    No Results Found