Estimating the 3D pose of desktop objects is crucial for applications such as robotic manipulation. Many existing approaches to this problem require a depth map of the object for both training and prediction, which restricts them to opaque, lambertian objects that produce good returns in an RGBD sensor. In this paper we forgo using a depth sensor in favor of raw stereo input. We address two problems: first, we establish an easy method for capturing and labeling 3D keypoints on desktop objects with an RGB camera; and second, we develop a deep neural network, called KeyPose, that learns to accurately predict object poses using 3D keypoints, from stereo input, and works even for transparent objects. To evaluate the performance of our method, we create a dataset of 15 clear objects in five classes, with 48K 3D-keypoint labeled images. We train both instance and category models, and show generalization to new textures, poses, and objects. KeyPose surpasses state-of-the-art performance in 3D pose estimation on this dataset by factors of 1.5 to 3.5, even in cases where the competing method is provided with ground-truth depth. Stereo input is essential for this performance as it improves results compared to using monocular input by a factor of 2. We will release a public version of the data capture and labeling pipeline, the transparent object database, and the KeyPose models and evaluation code. Project website: https://sites.google.com/corp/view/keypose.

Mapping and localization, preferably from a small number of observations, are fundamental tasks in robotics. We address these tasks by combining spatial structure (differentiable mapping) and end-to-end learning in a novel neural network architecture: the Differentiable Mapping Network (DMN). The DMN constructs a spatially structured view-embedding map and uses it for subsequent visual localization with a particle filter. Since the DMN architecture is end-to-end differentiable, we can jointly learn the map representation and localization using gradient descent. We apply the DMN to sparse visual localization, where a robot needs to localize in a new environment with respect to a small number of images from known viewpoints. We evaluate the DMN using simulated environments and a challenging real-world Street View dataset. We find that the DMN learns effective map representations for visual localization. The benefit of spatial structure increases with larger environments, more viewpoints for mapping, and when training data is scarce. Project website: https://sites.google.com/view/differentiable-mapping.

We present a novel method for simultaneous learning of depth, egomotion, object motion, and camera intrinsics from monocular videos, using only consistency across neighboring video frames as supervision signal. Similarly to prior work, our method learns by applying differentiable warping to frames and comparing the result to adjacent ones, but it provides several improvements: We address occlusions geometrically and differentiably, directly using the depth maps as predicted during training. We introduce randomized layer normalization, a novel powerful regularizer, and we account for object motion relative to the scene. To the best of our knowledge, our work is the first to learn the camera intrinsic parameters, including lens distortion, from video in an unsupervised manner, thereby allowing us to extract accurate depth and motion from arbitrary videos of unknown origin at scale. We evaluate our results on the Cityscapes, KITTI and EuRoC datasets, establishing new state of the art on depth prediction and odometry, and demonstrate qualitatively that depth prediction can be learned from a collection of YouTube videos. The code will be open sourced once anonymity is lifted.

We introduce Recurrent State Representation Learning (RSRL) to tackle the problem of state representation learning in robotics for partially observable environments. To learn low-dimensional state representations, we combine a Long Short Term Memory network with robotic priors. RSRL introduces new priors with landmarks and combines them with existing robotics priors from the literature to train the representations. To evaluate the quality of the learned state representation, we introduce validation networks that help us better visualize and quantitatively analyze the learned state representations. We show that the learned representations are low-dimensional, locally consistent, and can approximate the underlying true state for robot localization in simulated 3D maze environments. We use the learned representations for reinforcement learning and show that we achieve similar performance as training with the true state. The learned representations are robust to landmark misclassification errors.

We describe the winning entry to the Amazon Picking Challenge 2015. From the experience of building this system and competing, we derive several conclusions: (1) We suggest to characterize robotic system building along four key aspects, each of them spanning a spectrum of solutions - modularity vs. integration, generality vs. assumptions, computation vs. embodiment, and planning vs. feedback. (2) To understand which region of each spectrum most adequately addresses which robotic problem, we must explore the full spectrum of possible approaches. (3) For manipulation problems in unstructured environments, certain regions of each spectrum match the problem most adequately, and should be exploited further. This is supported by the fact that our solution deviated from the majority of the other challenge entries along each of the spectra. This is an abridged version of a conference publication.

We present a method for multi-class segmentation from RGB-D data in a realistic warehouse picking setting. The method computes pixel-wise probabilities and combines them to find a coherent object segmentation. It reliably segments objects in cluttered scenarios, even when objects are translucent, reflective, highly deformable, have fuzzy surfaces, or consist of loosely coupled components. The robust performance results from the exploitation of problem structure inherent to the warehouse setting. The proposed method proved its capabilities as part of our winning entry to the 2015 Amazon Picking Challenge. We present a detailed experimental analysis of the contribution of different information sources, compare our method to standard segmentation techniques, and assess possible extensions that further enhance the algorithm's capabilities. We release our software and data sets as open source.