Gregory D. Hager

Jan Emily Mangulabnan, Akshat Chauhan, Laura Fleig et al.

In endoscopic surgery, surgeons continuously locate the endoscopic view relative to the anatomy by interpreting the evolving visual appearance of the intraoperative scene in the context of their prior knowledge. Vision-based navigation systems seek to replicate this capability by recovering camera pose directly from endoscopic video, but most approaches do not embody the same principles of reasoning about new frames that makes surgeons successful. Instead, they remain grounded in feature matching and geometric optimization over keyframes, an approach that has been shown to degrade under the challenging conditions of endoscopic imaging like low texture and rapid illumination changes. Here, we pursue an alternative approach and investigate a policy-based formulation of endoscopic camera pose recovery that seeks to imitate experts in estimating trajectories conditioned on the previous camera state. Our approach directly predicts short-horizon relative motions without maintaining an explicit geometric representation at inference time. It thus addresses, by design, some of the notorious challenges of geometry-based approaches, such as brittle correspondence matching, instability in texture-sparse regions, and limited pose coverage due to reconstruction failure. We evaluate the proposed formulation on cadaveric sinus endoscopy. Under oracle state conditioning, we compare short-horizon motion prediction quality to geometric baselines achieving lowest mean translation error and competitive rotational accuracy. We analyze robustness by grouping prediction windows according to texture richness and illumination change indicating reduced sensitivity to low-texture conditions. These findings suggest that a learned motion policy offers a viable alternative formulation for endoscopic camera pose recovery.

Russell H. Taylor, Gregory D. Hager, Ralph Etienne-Cummings. Eric Grimson et al.

In the last ten years, medical robotics has moved from the margins to the mainstream. Since the Engineering Research Center for Computer-Integrated Surgical Systems and Technology was Launched in 1998 with National Science Foundation funding, medical robots have been promoted from handling routine tasks to performing highly sophisticated interventions and related assignments. The CISST ERC has played a significant role in this transformation. And thanks to NSF support, the ERC has built the professional infrastructure that will continue our mission: bringing data and technology together in clinical systems that will dramatically change how surgery and other procedures are done. The enhancements we envision touch virtually every aspect of the delivery of care: - More accurate procedures - More consistent, predictable results from one patient to the next - Improved clinical outcomes - Greater patient safety - Reduced liability for healthcare providers - Lower costs for everyone - patients, facilities, insurers, government - Easier, faster recovery for patients - Effective new ways to treat health problems - Healthier patients, and a healthier system The basic science and engineering the ERC is developing now will yield profound benefits for all concerned about health care - from government agencies to insurers, from clinicians to patients to the general public. All will experience the healing touch of medical robotics, thanks in no small part to the work of the CISST ERC and its successors.

Xingtong Liu, Zhaoshuo Li, Masaru Ishii et al.

In endoscopy, many applications (e.g., surgical navigation) would benefit from a real-time method that can simultaneously track the endoscope and reconstruct the dense 3D geometry of the observed anatomy from a monocular endoscopic video. To this end, we develop a Simultaneous Localization and Mapping system by combining the learning-based appearance and optimizable geometry priors and factor graph optimization. The appearance and geometry priors are explicitly learned in an end-to-end differentiable training pipeline to master the task of pair-wise image alignment, one of the core components of the SLAM system. In our experiments, the proposed SLAM system is shown to robustly handle the challenges of texture scarceness and illumination variation that are commonly seen in endoscopy. The system generalizes well to unseen endoscopes and subjects and performs favorably compared with a state-of-the-art feature-based SLAM system. The code repository is available at https://github.com/lppllppl920/SAGE-SLAM.git.

David Z. Li, Masaru Ishii, Russell H. Taylor et al.

In this work, we explore whether it is possible to learn representations of endoscopic video frames to perform tasks such as identifying surgical tool presence without supervision. We use a maximum mean discrepancy (MMD) variational autoencoder (VAE) to learn low-dimensional latent representations of endoscopic videos and manipulate these representations to distinguish frames containing tools from those without tools. We use three different methods to manipulate these latent representations in order to predict tool presence in each frame. Our fully unsupervised methods can identify whether endoscopic video frames contain tools with average precision of 71.56, 73.93, and 76.18, respectively, comparable to supervised methods. Our code is available at https://github.com/zdavidli/tool-presence/

Xingtong Liu, Maia Stiber, Jindan Huang et al.

Reconstructing accurate 3D surface models of sinus anatomy directly from an endoscopic video is a promising avenue for cross-sectional and longitudinal analysis to better understand the relationship between sinus anatomy and surgical outcomes. We present a patient-specific, learning-based method for 3D reconstruction of sinus surface anatomy directly and only from endoscopic videos. We demonstrate the effectiveness and accuracy of our method on in and ex vivo data where we compare to sparse reconstructions from Structure from Motion, dense reconstruction from COLMAP, and ground truth anatomy from CT. Our textured reconstructions are watertight and enable measurement of clinically relevant parameters in good agreement with CT. The source code is available at https://github.com/lppllppl920/DenseReconstruction-Pytorch.

Xingtong Liu, Yiping Zheng, Benjamin Killeen et al.

High-quality 3D reconstructions from endoscopy video play an important role in many clinical applications, including surgical navigation where they enable direct video-CT registration. While many methods exist for general multi-view 3D reconstruction, these methods often fail to deliver satisfactory performance on endoscopic video. Part of the reason is that local descriptors that establish pair-wise point correspondences, and thus drive reconstruction, struggle when confronted with the texture-scarce surface of anatomy. Learning-based dense descriptors usually have larger receptive fields enabling the encoding of global information, which can be used to disambiguate matches. In this work, we present an effective self-supervised training scheme and novel loss design for dense descriptor learning. In direct comparison to recent local and dense descriptors on an in-house sinus endoscopy dataset, we demonstrate that our proposed dense descriptor can generalize to unseen patients and scopes, thereby largely improving the performance of Structure from Motion (SfM) in terms of model density and completeness. We also evaluate our method on a public dense optical flow dataset and a small-scale SfM public dataset to further demonstrate the effectiveness and generality of our method. The source code is available at https://github.com/lppllppl920/DenseDescriptorLearning-Pytorch.

Michael Peven, Gregory D. Hager, Austin Reiter

Traditional action recognition models are constructed around the paradigm of 2D perspective imagery. Though sophisticated time-series models have pushed the field forward, much of the information is still not exploited by confining the domain to 2D. In this work, we introduce a novel representation of motion as a voxelized 3D vector field and demonstrate how it can be used to improve performance of action recognition networks. This volumetric representation is a natural fit for 3D CNNs, and allows out-of-plane data augmentation techniques during training of these networks. Both the construction of this representation from RGB-D video and inference can be run in real time. We demonstrate superior results using this representation with our network design on the open-source NTU RGB+D dataset where it outperforms state-of-the-art on both of the defined evaluation metrics. Furthermore, we experimentally show how the out-of-plane augmentation techniques create viewpoint invariance and allow the model trained using this representation to generalize to unseen camera angles. Code is available here: https://github.com/mpeven/ntu_rgb.

Andrew Hundt, Benjamin Killeen, Nicholas Greene et al.

Current Reinforcement Learning (RL) algorithms struggle with long-horizon tasks where time can be wasted exploring dead ends and task progress may be easily reversed. We develop the SPOT framework, which explores within action safety zones, learns about unsafe regions without exploring them, and prioritizes experiences that reverse earlier progress to learn with remarkable efficiency. The SPOT framework successfully completes simulated trials of a variety of tasks, improving a baseline trial success rate from 13% to 100% when stacking 4 cubes, from 13% to 99% when creating rows of 4 cubes, and from 84% to 95% when clearing toys arranged in adversarial patterns. Efficiency with respect to actions per trial typically improves by 30% or more, while training takes just 1-20k actions, depending on the task. Furthermore, we demonstrate direct sim to real transfer. We are able to create real stacks in 100% of trials with 61% efficiency and real rows in 100% of trials with 59% efficiency by directly loading the simulation-trained model on the real robot with no additional real-world fine-tuning. To our knowledge, this is the first instance of reinforcement learning with successful sim to real transfer applied to long term multi-step tasks such as block-stacking and row-making with consideration of progress reversal. Code is available at https://github.com/jhu-lcsr/good_robot .

Xingtong Liu, Ayushi Sinha, Masaru Ishii et al.

We present a self-supervised approach to training convolutional neural networks for dense depth estimation from monocular endoscopy data without a priori modeling of anatomy or shading. Our method only requires monocular endoscopic videos and a multi-view stereo method, e.g., structure from motion, to supervise learning in a sparse manner. Consequently, our method requires neither manual labeling nor patient computed tomography (CT) scan in the training and application phases. In a cross-patient experiment using CT scans as groundtruth, the proposed method achieved submillimeter mean residual error. In a comparison study to recent self-supervised depth estimation methods designed for natural video on in vivo sinus endoscopy data, we demonstrate that the proposed approach outperforms the previous methods by a large margin. The source code for this work is publicly available online at https://github.com/lppllppl920/EndoscopyDepthEstimation-Pytorch.

Ayushi Sinha, Masaru Ishii, Russell H. Taylor et al.

Registering images from different modalities is an active area of research in computer aided medical interventions. Several registration algorithms have been developed, many of which achieve high accuracy. However, these results are dependent on many factors, including the quality of the extracted features or segmentations being registered as well as the initial alignment. Although several methods have been developed towards improving segmentation algorithms and automating the segmentation process, few automatic initialization algorithms have been explored. In many cases, the initial alignment from which a registration is initiated is performed manually, which interferes with the clinical workflow. Our aim is to use scene classification in endoscopic procedures to achieve coarse alignment of the endoscope and a preoperative image of the anatomy. In this paper, we show using simulated scenes that a neural network can predict the region of anatomy (with respect to a preoperative image) that the endoscope is located in by observing a single endoscopic video frame. With limited training and without any hyperparameter tuning, our method achieves an accuracy of 76.53 (+/-1.19)%. There are several avenues for improvement, making this a promising direction of research. Code is available at https://github.com/AyushiSinha/AutoInitialization.

Wentao Zhu, Xiang Xiang, Trac D. Tran et al.

Mass segmentation provides effective morphological features which are important for mass diagnosis. In this work, we propose a novel end-to-end network for mammographic mass segmentation which employs a fully convolutional network (FCN) to model a potential function, followed by a CRF to perform structured learning. Because the mass distribution varies greatly with pixel position, the FCN is combined with a position priori. Further, we employ adversarial training to eliminate over-fitting due to the small sizes of mammogram datasets. Multi-scale FCN is employed to improve the segmentation performance. Experimental results on two public datasets, INbreast and DDSM-BCRP, demonstrate that our end-to-end network achieves better performance than state-of-the-art approaches. \footnote{https://github.com/wentaozhu/adversarial-deep-structural-networks.git}

Robert DiPietro, Colin Lea, Anand Malpani et al.

We apply recurrent neural networks to the task of recognizing surgical activities from robot kinematics. Prior work in this area focuses on recognizing short, low-level activities, or gestures, and has been based on variants of hidden Markov models and conditional random fields. In contrast, we work on recognizing both gestures and longer, higher-level activites, or maneuvers, and we model the mapping from kinematics to gestures/maneuvers with recurrent neural networks. To our knowledge, we are the first to apply recurrent neural networks to this task. Using a single model and a single set of hyperparameters, we match state-of-the-art performance for gesture recognition and advance state-of-the-art performance for maneuver recognition, in terms of both accuracy and edit distance. Code is available at https://github.com/rdipietro/miccai-2016-surgical-activity-rec .

Weiyao Wang, Marin Kobilarov, Gregory D. Hager

Recent advances in latent space dynamics model from pixels show promising progress in vision-based model predictive control (MPC). However, executing MPC in real time can be challenging due to its intensive computational cost in each timestep. We propose to introduce additional learning objectives to enforce that the learned latent space is proportional derivative controllable. In execution time, the simple PD-controller can be applied directly to the latent space encoded from pixels, to produce simple and effective control to systems with visual observations. We show that our method outperforms baseline methods to produce robust goal reaching and trajectory tracking in various environments.

Will Pryor, Yotam Barnoy, Suraj Raval et al.

Real-time visual localization of needles is necessary for various surgical applications, including surgical automation and visual feedback. In this study we investigate localization and autonomous robotic control of needles in the context of our magneto-suturing system. Our system holds the potential for surgical manipulation with the benefit of minimal invasiveness and reduced patient side effects. However, the non-linear magnetic fields produce unintuitive forces and demand delicate position-based control that exceeds the capabilities of direct human manipulation. This makes automatic needle localization a necessity. Our localization method combines neural network-based segmentation and classical techniques, and we are able to consistently locate our needle with 0.73 mm RMS error in clean environments and 2.72 mm RMS error in challenging environments with blood and occlusion. The average localization RMS error is 2.16 mm for all environments we used in the experiments. We combine this localization method with our closed-loop feedback control system to demonstrate the further applicability of localization to autonomous control. Our needle is able to follow a running suture path in (1) no blood, no tissue; (2) heavy blood, no tissue; (3) no blood, with tissue; and (4) heavy blood, with tissue environments. The tip position tracking error ranges from 2.6 mm to 3.7 mm RMS, opening the door towards autonomous suturing tasks.

Gordon Christie, Kevin Foster, Shea Hagstrom et al.

Current methods for Earth observation tasks such as semantic mapping, map alignment, and change detection rely on near-nadir images; however, often the first available images in response to dynamic world events such as natural disasters are oblique. These tasks are much more difficult for oblique images due to observed object parallax. There has been recent success in learning to regress geocentric pose, defined as height above ground and orientation with respect to gravity, by training with airborne lidar registered to satellite images. We present a model for this novel task that exploits affine invariance properties to outperform state of the art performance by a wide margin. We also address practical issues required to deploy this method in the wild for real-world applications. Our data and code are publicly available.

Kapil D. Katyal, I-Jeng Wang, Gregory D. Hager

Accurate state and uncertainty estimation is imperative for mobile robots and self driving vehicles to achieve safe navigation in pedestrian rich environments. A critical component of state and uncertainty estimation for robot navigation is to perform robustly under out-of-distribution noise. Traditional methods of state estimation decouple perception and state estimation making it difficult to operate on noisy, high dimensional data. Here, we describe an approach that combines the expressiveness of deep neural networks with principled approaches to uncertainty estimation found in recursive filters. We particularly focus on techniques that provide better robustness to out-of-distribution noise and demonstrate applicability of our approach on two scenarios: a simple noisy pendulum state estimation problem and real world pedestrian localization using the nuScenes dataset. We show that our approach improves state and uncertainty estimation compared to baselines while achieving approximately 3x improvement in computational efficiency.

Tae Soo Kim, Jonathan Jones, Gregory D. Hager

We present a dual-pathway approach for recognizing fine-grained interactions from videos. We build on the success of prior dual-stream approaches, but make a distinction between the static and dynamic representations of objects and their interactions explicit by introducing separate motion and object detection pathways. Then, using our new Motion-Guided Attention Fusion module, we fuse the bottom-up features in the motion pathway with features captured from object detections to learn the temporal aspects of an action. We show that our approach can generalize across appearance effectively and recognize actions where an actor interacts with previously unseen objects. We validate our approach using the compositional action recognition task from the Something-Something-v2 dataset where we outperform existing state-of-the-art methods. We also show that our method can generalize well to real world tasks by showing state-of-the-art performance on recognizing humans assembling various IKEA furniture on the IKEA-ASM dataset.

Daniel Neimark, Omri Bar, Maya Zohar et al.

Prior work demonstrated the ability of machine learning to automatically recognize surgical workflow steps from videos. However, these studies focused on only a single type of procedure. In this work, we analyze, for the first time, surgical step recognition on four different laparoscopic surgeries: Cholecystectomy, Right Hemicolectomy, Sleeve Gastrectomy, and Appendectomy. Inspired by the traditional apprenticeship model, in which surgical training is based on the Halstedian method, we paraphrase the "see one, do one, teach one" approach for the surgical intelligence domain as "train one, classify one, teach one". In machine learning, this approach is often referred to as transfer learning. To analyze the impact of transfer learning across different laparoscopic procedures, we explore various time-series architectures and examine their performance on each target domain. We introduce a new architecture, the Time-Series Adaptation Network (TSAN), an architecture optimized for transfer learning of surgical step recognition, and we show how TSAN can be pre-trained using self-supervised learning on a Sequence Sorting task. Such pre-training enables TSAN to learn workflow steps of a new laparoscopic procedure type from only a small number of labeled samples from the target procedure. Our proposed architecture leads to better performance compared to other possible architectures, reaching over 90% accuracy when transferring from laparoscopic Cholecystectomy to the other three procedure types.

Alaa Eldin Abdelaal, Nancy Hong, Apeksha Avinash et al.

We propose a new method for six-degree-of-freedom (6-DoF) autonomous camera movement for minimally invasive surgery, which, unlike previous methods, takes into account both the position and orientation information from structures in the surgical scene. In addition to locating the camera for a good view of the manipulated object, our autonomous camera takes into account workspace constraints, including the horizon and safety constraints. We developed a simulation environment to test our method on the "wire chaser" surgical training task from validated training curricula in conventional laparoscopy and robot-assisted surgery. Furthermore, we propose, for the first time, the application of the proposed autonomous camera method in video-based surgical skill assessment, an area where videos are typically recorded using fixed cameras. In a study with N=30 human subjects, we show that video examination of the autonomous camera view as it tracks the ring motion over the wire leads to more accurate user error (ring touching the wire) detection than when using a fixed camera view, or camera movement with a fixed orientation. Our preliminary work suggests that there are potential benefits to autonomous camera positioning informed by scene orientation, and this can direct designers of automated endoscopes and surgical robotic systems, especially when using chip-on-tip cameras that can be wristed for 6-DoF motion.

Tae Soo Kim, Gregory D. Hager

We present a general framework for compositional action recognition -- i.e. action recognition where the labels are composed out of simpler components such as subjects, atomic-actions and objects. The main challenge in compositional action recognition is that there is a combinatorially large set of possible actions that can be composed using basic components. However, compositionality also provides a structure that can be exploited. To do so, we develop and test a novel Structured Attention Fusion (SAF) self-attention mechanism to combine information from object detections, which capture the time-series structure of an action, with visual cues that capture contextual information. We show that our approach recognizes novel verb-noun compositions more effectively than current state of the art systems, and it generalizes to unseen action categories quite efficiently from only a few labeled examples. We validate our approach on the challenging Something-Else tasks from the Something-Something-V2 dataset. We further show that our framework is flexible and can generalize to a new domain by showing competitive results on the Charades-Fewshot dataset.

Jonathan D. Jones, Cathryn Cortesa, Amy Shelton et al.

In this paper we address the task of recognizing assembly actions as a structure (e.g. a piece of furniture or a toy block tower) is built up from a set of primitive objects. Recognizing the full range of assembly actions requires perception at a level of spatial detail that has not been attempted in the action recognition literature to date. We extend the fine-grained activity recognition setting to address the task of assembly action recognition in its full generality by unifying assembly actions and kinematic structures within a single framework. We use this framework to develop a general method for recognizing assembly actions from observation sequences, along with observation features that take advantage of a spatial assembly's special structure. Finally, we evaluate our method empirically on two application-driven data sources: (1) An IKEA furniture-assembly dataset, and (2) A block-building dataset. On the first, our system recognizes assembly actions with an average framewise accuracy of 70% and an average normalized edit distance of 10%. On the second, which requires fine-grained geometric reasoning to distinguish between assemblies, our system attains an average normalized edit distance of 23% -- a relative improvement of 69% over prior work.

Qihao Liu, Weichao Qiu, Weiyao Wang et al.

We propose an unsupervised vision-based system to estimate the joint configurations of the robot arm from a sequence of RGB or RGB-D images without knowing the model a priori, and then adapt it to the task of category-independent articulated object pose estimation. We combine a classical geometric formulation with deep learning and extend the use of epipolar constraint to multi-rigid-body systems to solve this task. Given a video sequence, the optical flow is estimated to get the pixel-wise dense correspondences. After that, the 6D pose is computed by a modified PnP algorithm. The key idea is to leverage the geometric constraints and the constraint between multiple frames. Furthermore, we build a synthetic dataset with different kinds of robots and multi-joint articulated objects for the research of vision-based robot control and robotic vision. We demonstrate the effectiveness of our method on three benchmark datasets and show that our method achieves higher accuracy than the state-of-the-art supervised methods in estimating joint angles of robot arms and articulated objects.

Ji Woong Kim, Changyan He, Muller Urias et al.

A fundamental challenge in retinal surgery is safely navigating a surgical tool to a desired goal position on the retinal surface while avoiding damage to surrounding tissues, a procedure that typically requires tens-of-microns accuracy. In practice, the surgeon relies on depth-estimation skills to localize the tool-tip with respect to the retina in order to perform the tool-navigation task, which can be prone to human error. To alleviate such uncertainty, prior work has introduced ways to assist the surgeon by estimating the tool-tip distance to the retina and providing haptic or auditory feedback. However, automating the tool-navigation task itself remains unsolved and largely unexplored. Such a capability, if reliably automated, could serve as a building block to streamline complex procedures and reduce the chance for tissue damage. Towards this end, we propose to automate the tool-navigation task by learning to mimic expert demonstrations of the task. Specifically, a deep network is trained to imitate expert trajectories toward various locations on the retina based on recorded visual servoing to a given goal specified by the user. The proposed autonomous navigation system is evaluated in simulation and in physical experiments using a silicone eye phantom. We show that the network can reliably navigate a needle surgical tool to various desired locations within 137 microns accuracy in physical experiments and 94 microns in simulation on average, and generalizes well to unseen situations such as in the presence of auxiliary surgical tools, variable eye backgrounds, and brightness conditions.

Lena Maier-Hein, Matthias Eisenmann, Duygu Sarikaya et al.

Recent developments in data science in general and machine learning in particular have transformed the way experts envision the future of surgery. Surgical Data Science (SDS) is a new research field that aims to improve the quality of interventional healthcare through the capture, organization, analysis and modeling of data. While an increasing number of data-driven approaches and clinical applications have been studied in the fields of radiological and clinical data science, translational success stories are still lacking in surgery. In this publication, we shed light on the underlying reasons and provide a roadmap for future advances in the field. Based on an international workshop involving leading researchers in the field of SDS, we review current practice, key achievements and initiatives as well as available standards and tools for a number of topics relevant to the field, namely (1) infrastructure for data acquisition, storage and access in the presence of regulatory constraints, (2) data annotation and sharing and (3) data analytics. We further complement this technical perspective with (4) a review of currently available SDS products and the translational progress from academia and (5) a roadmap for faster clinical translation and exploitation of the full potential of SDS, based on an international multi-round Delphi process.

Haomin Chen, Shun Miao, Daguang Xu et al.

CXRs are a crucial and extraordinarily common diagnostic tool, leading to heavy research for CAD solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep HMLC approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. Finally, we demonstrate that HMLC can be an effective means to manage missing or incomplete labels. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting abnormality labels from the CXR arm of the PLCO dataset, which comprises over $198,000$ manually annotated CXRs. When using complete labels, we report a mean AUC of 0.887, the highest yet reported for this dataset. These results are supported by ancillary experiments on the PadChest dataset, where we also report significant improvements, 1.2% and 4.1% in AUC and AP, respectively over strong "flat" classifiers. Finally, we demonstrate that our HMLC approach can much better handle incompletely labelled data. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.

Haomin Chen, Yirui Wang, Kang Zheng et al.

Visual cues of enforcing bilaterally symmetric anatomies as normal findings are widely used in clinical practice to disambiguate subtle abnormalities from medical images. So far, inadequate research attention has been received on effectively emulating this practice in CAD methods. In this work, we exploit semantic anatomical symmetry or asymmetry analysis in a complex CAD scenario, i.e., anterior pelvic fracture detection in trauma PXRs, where semantically pathological (refer to as fracture) and non-pathological (e.g., pose) asymmetries both occur. Visually subtle yet pathologically critical fracture sites can be missed even by experienced clinicians, when limited diagnosis time is permitted in emergency care. We propose a novel fracture detection framework that builds upon a Siamese network enhanced with a spatial transformer layer to holistically analyze symmetric image features. Image features are spatially formatted to encode bilaterally symmetric anatomies. A new contrastive feature learning component in our Siamese network is designed to optimize the deep image features being more salient corresponding to the underlying semantic asymmetries (caused by pelvic fracture occurrences). Our proposed method have been extensively evaluated on 2,359 PXRs from unique patients (the largest study to-date), and report an area under ROC curve score of 0.9771. This is the highest among state-of-the-art fracture detection methods, with improved clinical indications.

Gordon Christie, Rodrigo Rene Rai Munoz Abujder, Kevin Foster et al.

An object's geocentric pose, defined as the height above ground and orientation with respect to gravity, is a powerful representation of real-world structure for object detection, segmentation, and localization tasks using RGBD images. For close-range vision tasks, height and orientation have been derived directly from stereo-computed depth and more recently from monocular depth predicted by deep networks. For long-range vision tasks such as Earth observation, depth cannot be reliably estimated with monocular images. Inspired by recent work in monocular height above ground prediction and optical flow prediction from static images, we develop an encoding of geocentric pose to address this challenge and train a deep network to compute the representation densely, supervised by publicly available airborne lidar. We exploit these attributes to rectify oblique images and remove observed object parallax to dramatically improve the accuracy of localization and to enable accurate alignment of multiple images taken from very different oblique viewpoints. We demonstrate the value of our approach by extending two large-scale public datasets for semantic segmentation in oblique satellite images. All of our data and code are publicly available.

Helisa Dhamo, Azade Farshad, Iro Laina et al.

Image manipulation can be considered a special case of image generation where the image to be produced is a modification of an existing image. Image generation and manipulation have been, for the most part, tasks that operate on raw pixels. However, the remarkable progress in learning rich image and object representations has opened the way for tasks such as text-to-image or layout-to-image generation that are mainly driven by semantics. In our work, we address the novel problem of image manipulation from scene graphs, in which a user can edit images by merely applying changes in the nodes or edges of a semantic graph that is generated from the image. Our goal is to encode image information in a given constellation and from there on generate new constellations, such as replacing objects or even changing relationships between objects, while respecting the semantics and style from the original image. We introduce a spatio-semantic scene graph network that does not require direct supervision for constellation changes or image edits. This makes it possible to train the system from existing real-world datasets with no additional annotation effort.

Pengfei Li, Weichao Qiu, Michael Peven et al.

Scene context is a powerful constraint on the geometry of objects within the scene in cases, such as surveillance, where the camera geometry is unknown and image quality may be poor. In this paper, we describe a method for estimating the pose of cars in a scene jointly with the ground plane that supports them. We formulate this as a joint optimization that accounts for varying car shape using a statistical atlas, and which simultaneously computes geometry and internal camera parameters. We demonstrate that this method produces significant improvements for car pose estimation, and we show that the resulting 3D geometry, when computed over a video sequence, makes it possible to improve on state of the art classification of car behavior. We also show that introducing the planar constraint allows us to estimate camera focal length in a reliable manner.

Tae Soo Kim, Jonathan D. Jones, Michael Peven et al.

There are many realistic applications of activity recognition where the set of potential activity descriptions is combinatorially large. This makes end-to-end supervised training of a recognition system impractical as no training set is practically able to encompass the entire label set. In this paper, we present an approach to fine-grained recognition that models activities as compositions of dynamic action signatures. This compositional approach allows us to reframe fine-grained recognition as zero-shot activity recognition, where a detector is composed "on the fly" from simple first-principles state machines supported by deep-learned components. We evaluate our method on the Olympic Sports and UCF101 datasets, where our model establishes a new state of the art under multiple experimental paradigms. We also extend this method to form a unique framework for zero-shot joint segmentation and classification of activities in video and demonstrate the first results in zero-shot decoding of complex action sequences on a widely-used surgical dataset. Lastly, we show that we can use off-the-shelf object detectors to recognize activities in completely de-novo settings with no additional training.

Yi Zhang, Xinyue Wei, Weichao Qiu et al.

Despite the rapid growth in datasets for video activity, stable robust activity recognition with neural networks remains challenging. This is in large part due to the explosion of possible variation in video -- including lighting changes, object variation, movement variation, and changes in surrounding context. An alternative is to make use of simulation data, where all of these factors can be artificially controlled. In this paper, we propose the Randomized Simulation as Augmentation (RSA) framework which augments real-world training data with synthetic data to improve the robustness of action recognition networks. We generate large-scale synthetic datasets with randomized nuisance factors. We show that training with such extra data, when appropriately constrained, can significantly improve the performance of the state-of-the-art I3D networks or, conversely, reduce the number of labeled real videos needed to achieve good performance. Experiments on two real-world datasets NTU RGB+D and VIRAT demonstrate the effectiveness of our method.

Xingtong Liu, Ayushi Sinha, Masaru Ishii et al.

We present a self-supervised learning-based pipeline for dense 3D reconstruction from full-length monocular endoscopic videos without a priori modeling of anatomy or shading. Our method only relies on unlabeled monocular endoscopic videos and conventional multi-view stereo algorithms, and requires neither manual interaction nor patient CT in both training and application phases. In a cross-patient study using CT scans as groundtruth, we show that our method is able to produce photo-realistic dense 3D reconstructions with submillimeter mean residual errors from endoscopic videos from unseen patients and scopes.

Robert DiPietro, Gregory D. Hager

Prior work has demonstrated the feasibility of automated activity recognition in robot-assisted surgery from motion data. However, these efforts have assumed the availability of a large number of densely-annotated sequences, which must be provided manually by experts. This process is tedious, expensive, and error-prone. In this paper, we present the first analysis under the assumption of scarce annotations, where as little as one annotated sequence is available for training. We demonstrate feasibility of automated recognition in this challenging setting, and we show that learning representations in an unsupervised fashion, before the recognition phase, leads to significant gains in performance. In addition, our paper poses a new challenge to the community: how much further can we push performance in this important yet relatively unexplored regime?

Andrew Hundt, Varun Jain, Gregory D. Hager

Neural Architecture Search (NAS) has been a source of dramatic improvements in neural network design, with recent results meeting or exceeding the performance of hand-tuned architectures. However, our understanding of how to represent the search space for neural net architectures and how to search that space efficiently are both still in their infancy. We have performed an in-depth analysis to identify limitations in a widely used search space and a recent architecture search method, Differentiable Architecture Search (DARTS). These findings led us to introduce novel network blocks with a more general, balanced, and consistent design; a better-optimized Cosine Power Annealing learning rate schedule; and other improvements. Our resulting sharpDARTS search is 50% faster with a 20-30% relative improvement in final model error on CIFAR-10 when compared to DARTS. Our best single model run has 1.93% (1.98+/-0.07) validation error on CIFAR-10 and 5.5% error (5.8+/-0.3) on the recently released CIFAR-10.1 test set. To our knowledge, both are state of the art for models of similar size. This model also generalizes competitively to ImageNet at 25.1% top-1 (7.8% top-5) error. We found improvements for existing search spaces but does DARTS generalize to new domains? We propose Differentiable Hyperparameter Grid Search and the HyperCuboid search space, which are representations designed to leverage DARTS for more general parameter optimization. Here we find that DARTS fails to generalize when compared against a human's one shot choice of models. We look back to the DARTS and sharpDARTS search spaces to understand why, and an ablation study reveals an unusual generalization gap. We finally propose Max-W regularization to solve this problem, which proves significantly better than the handmade design. Code will be made available.

Chris Paxton, Felix Jonathan, Andrew Hundt et al.

How can we enable users to create effective, perception-driven task plans for collaborative robots? We conducted a 35-person user study with the Behavior Tree-based CoSTAR system to determine which strategies for end user creation of generalizable robot task plans are most usable and effective. CoSTAR allows domain experts to author complex, perceptually grounded task plans for collaborative robots. As a part of CoSTAR's wide range of capabilities, it allows users to specify SmartMoves: abstract goals such as "pick up component A from the right side of the table." Users were asked to perform pick-and-place assembly tasks with either SmartMoves or one of three simpler baseline versions of CoSTAR. Overall, participants found CoSTAR to be highly usable, with an average System Usability Scale score of 73.4 out of 100. SmartMove also helped users perform tasks faster and more effectively; all SmartMove users completed the first two tasks, while not all users completed the tasks using the other strategies. SmartMove users showed better performance for incorporating perception across all three tasks.

Andrew Hundt, Varun Jain, Chia-Hung Lin et al.

A robot can now grasp an object more effectively than ever before, but once it has the object what happens next? We show that a mild relaxation of the task and workspace constraints implicit in existing object grasping datasets can cause neural network based grasping algorithms to fail on even a simple block stacking task when executed under more realistic circumstances. To address this, we introduce the JHU CoSTAR Block Stacking Dataset (BSD), where a robot interacts with 5.1 cm colored blocks to complete an order-fulfillment style block stacking task. It contains dynamic scenes and real time-series data in a less constrained environment than comparable datasets. There are nearly 12,000 stacking attempts and over 2 million frames of real data. We discuss the ways in which this dataset provides a valuable resource for a broad range of other topics of investigation. We find that hand-designed neural networks that work on prior datasets do not generalize to this task. Thus, to establish a baseline for this dataset, we demonstrate an automated search of neural network based models using a novel multiple-input HyperTree MetaModel, and find a final model which makes reasonable 3D pose predictions for grasping and stacking on our dataset. The CoSTAR BSD, code, and instructions are available at https://sites.google.com/site/costardataset.

Robert DiPietro, Gregory D. Hager

We show that it is possible to learn meaningful representations of surgical motion, without supervision, by learning to predict the future. An architecture that combines an RNN encoder-decoder and mixture density networks (MDNs) is developed to model the conditional distribution over future motion given past motion. We show that the learned encodings naturally cluster according to high-level activities, and we demonstrate the usefulness of these learned encodings in the context of information retrieval, where a database of surgical motion is searched for suturing activity using a motion-based query. Future prediction with MDNs is found to significantly outperform simpler baselines as well as the best previously-published result for this task, advancing state-of-the-art performance from an F1 score of 0.60 +- 0.14 to 0.77 +- 0.05.

Ayushi Sinha, Xingtong Liu, Austin Reiter et al.

Clinical examinations that involve endoscopic exploration of the nasal cavity and sinuses often do not have a reference image to provide structural context to the clinician. In this paper, we present a system for navigation during clinical endoscopic exploration in the absence of computed tomography (CT) scans by making use of shape statistics from past CT scans. Using a deformable registration algorithm along with dense reconstructions from video, we show that we are able to achieve submillimeter registrations in in-vivo clinical data and are able to assign confidence to these registrations using confidence criteria established using simulated data.

Chris Paxton, Yotam Barnoy, Kapil Katyal et al.

Prospection, the act of predicting the consequences of many possible futures, is intrinsic to human planning and action, and may even be at the root of consciousness. Surprisingly, this idea has been explored comparatively little in robotics. In this work, we propose a neural network architecture and associated planning algorithm that (1) learns a representation of the world useful for generating prospective futures after the application of high-level actions, (2) uses this generative model to simulate the result of sequences of high-level actions in a variety of environments, and (3) uses this same representation to evaluate these actions and perform tree search to find a sequence of high-level actions in a new environment. Models are trained via imitation learning on a variety of domains, including navigation, pick-and-place, and a surgical robotics task. Our approach allows us to visualize intermediate motion goals and learn to plan complex activity from visual information.

Christian Rupprecht, Iro Laina, Nassir Navab et al.

Interaction and collaboration between humans and intelligent machines has become increasingly important as machine learning methods move into real-world applications that involve end users. While much prior work lies at the intersection of natural language and vision, such as image captioning or image generation from text descriptions, less focus has been placed on the use of language to guide or improve the performance of a learned visual processing algorithm. In this paper, we explore methods to flexibly guide a trained convolutional neural network through user input to improve its performance during inference. We do so by inserting a layer that acts as a spatio-semantic guide into the network. This guide is trained to modify the network's activations, either directly via an energy minimization scheme or indirectly through a recurrent model that translates human language queries to interaction weights. Learning the verbal interaction is fully automatic and does not require manual text annotations. We evaluate the method on two datasets, showing that guiding a pre-trained network can improve performance, and provide extensive insights into the interaction between the guide and the CNN.

Chi Li, Jin Bai, Gregory D. Hager

One core challenge in object pose estimation is to ensure accurate and robust performance for large numbers of diverse foreground objects amidst complex background clutter. In this work, we present a scalable framework for accurately inferring six Degree-of-Freedom (6-DoF) pose for a large number of object classes from single or multiple views. To learn discriminative pose features, we integrate three new capabilities into a deep Convolutional Neural Network (CNN): an inference scheme that combines both classification and pose regression based on a uniform tessellation of the Special Euclidean group in three dimensions (SE(3)), the fusion of class priors into the training process via a tiled class map, and an additional regularization using deep supervision with an object mask. Further, an efficient multi-view framework is formulated to address single-view ambiguity. We show that this framework consistently improves the performance of the single-view network. We evaluate our method on three large-scale benchmarks: YCB-Video, JHUScene-50 and ObjectNet-3D. Our approach achieves competitive or superior performance over the current state-of-the-art methods.

Kapil Katyal, Katie Popek, Chris Paxton et al.

Fast, collision-free motion through unknown environments remains a challenging problem for robotic systems. In these situations, the robot's ability to reason about its future motion is often severely limited by sensor field of view (FOV). By contrast, biological systems routinely make decisions by taking into consideration what might exist beyond their FOV based on prior experience. In this paper, we present an approach for predicting occupancy map representations of sensor data for future robot motions using deep neural networks. We evaluate several deep network architectures, including purely generative and adversarial models. Testing on both simulated and real environments we demonstrated performance both qualitatively and quantitatively, with SSIM similarity measure up to 0.899. We showed that it is possible to make predictions about occupied space beyond the physical robot's FOV from simulated training data. In the future, this method will allow robots to navigate through unknown environments in a faster, safer manner.

Chi Li, M. Zeeshan Zia, Quoc-Huy Tran et al.

Recent data-driven approaches to scene interpretation predominantly pose inference as an end-to-end black-box mapping, commonly performed by a Convolutional Neural Network (CNN). However, decades of work on perceptual organization in both human and machine vision suggests that there are often intermediate representations that are intrinsic to an inference task, and which provide essential structure to improve generalization. In this work, we explore an approach for injecting prior domain structure into neural network training by supervising hidden layers of a CNN with intermediate concepts that normally are not observed in practice. We formulate a probabilistic framework which formalizes these notions and predicts improved generalization via this deep supervision method. One advantage of this approach is that we are able to train only from synthetic CAD renderings of cluttered scenes, where concept values can be extracted, but apply the results to real images. Our implementation achieves the state-of-the-art performance of 2D/3D keypoint localization and image classification on real image benchmarks, including KITTI, PASCAL VOC, PASCAL3D+, IKEA, and CIFAR100. We provide additional evidence that our approach outperforms alternative forms of supervision, such as multi-task networks.

Chris Paxton, Kapil Katyal, Christian Rupprecht et al.

Prospection is an important part of how humans come up with new task plans, but has not been explored in depth in robotics. Predicting multiple task-level is a challenging problem that involves capturing both task semantics and continuous variability over the state of the world. Ideally, we would combine the ability of machine learning to leverage big data for learning the semantics of a task, while using techniques from task planning to reliably generalize to new environment. In this work, we propose a method for learning a model encoding just such a representation for task planning. We learn a neural net that encodes the $k$ most likely outcomes from high level actions from a given world. Our approach creates comprehensible task plans that allow us to predict changes to the environment many time steps into the future. We demonstrate this approach via application to a stacking task in a cluttered environment, where the robot must select between different colored blocks while avoiding obstacles, in order to perform a task. We also show results on a simple navigation task. Our algorithm generates realistic image and pose predictions at multiple points in a given task.

Felix Jonathan, Chris Paxton, Gregory D. Hager

Accurate knowledge of object poses is crucial to successful robotic manipulation tasks, and yet most current approaches only work in laboratory settings. Noisy sensors and cluttered scenes interfere with accurate pose recognition, which is problematic especially when performing complex tasks involving object interactions. This is because most pose estimation algorithms focus only on estimating objects from a single frame, which means they lack continuity between frames. Further, they often do not consider resulting physical properties of the predicted scene such as intersecting objects or objects in unstable positions. In this work, we enhance the accuracy and stability of estimated poses for a whole scene by enforcing these physical constraints over time through the integration of a physics simulation. This allows us to accurately determine relationships between objects for a construction task. Scene parsing performance was evaluated on both simulated and real- world data. We apply our method to a real-world block stacking task, where the robot must build a tall tower of colored blocks.

Gregory D. Hager, Randal Bryant, Eric Horvitz et al.

Advances in Artificial Intelligence require progress across all of computer science.

Anand Malpani, S. Swaroop Vedula, Henry C. Lin et al.

With introduction of new technologies in the operating room like the da Vinci Surgical System, training surgeons to use them effectively and efficiently is crucial in the delivery of better patient care. Coaching by an expert surgeon is effective in teaching relevant technical skills, but current methods to deliver effective coaching are limited and not scalable. We present a virtual reality simulation-based framework for automated virtual coaching in surgical education. We implement our framework within the da Vinci Skills Simulator. We provide three coaching modes ranging from a hands-on teacher (continuous guidance) to a handsoff guide (assistance upon request). We present six teaching cues targeted at critical learning elements of a needle passing task, which are shown to the user based on the coaching mode. These cues are graphical overlays which guide the user, inform them about sub-par performance, and show relevant video demonstrations. We evaluated our framework in a pilot randomized controlled trial with 16 subjects in each arm. In a post-study questionnaire, participants reported high comprehension of feedback, and perceived improvement in performance. After three practice repetitions of the task, the control arm (independent learning) showed better motion efficiency whereas the experimental arm (received real-time coaching) had better performance of learning elements (as per the ACS Resident Skills Curriculum). We observed statistically higher improvement in the experimental group based on one of the metrics (related to needle grasp orientation). In conclusion, we developed an automated coach that provides real-time cues for surgical training and demonstrated its feasibility.

Chris Paxton, Felix Jonathan, Andrew Hundt et al.

How can we enable novice users to create effective task plans for collaborative robots? Must there be a tradeoff between generalizability and ease of use? To answer these questions, we conducted a user study with the CoSTAR system, which integrates perception and reasoning into a Behavior Tree-based task plan editor. In our study, we ask novice users to perform simple pick-and-place assembly tasks under varying perception and planning capabilities. Our study shows that users found Behavior Trees to be an effective way of specifying task plans. Furthermore, users were also able to more quickly, effectively, and generally author task plans with the addition of CoSTAR's planning, perception, and reasoning capabilities. Despite these improvements, concepts associated with these capabilities were rated by users as less usable, and our results suggest a direction for further refinement.

Chris Paxton, Vasumathi Raman, Gregory D. Hager et al.

We consider task and motion planning in complex dynamic environments for problems expressed in terms of a set of Linear Temporal Logic (LTL) constraints, and a reward function. We propose a methodology based on reinforcement learning that employs deep neural networks to learn low-level control policies as well as task-level option policies. A major challenge in this setting, both for neural network approaches and classical planning, is the need to explore future worlds of a complex and interactive environment. To this end, we integrate Monte Carlo Tree Search with hierarchical neural net control policies trained on expressive LTL specifications. This paper investigates the ability of neural networks to learn both LTL constraints and control policies in order to generate task plans in complex environments. We demonstrate our approach in a simulated autonomous driving setting, where a vehicle must drive down a road in traffic, avoid collisions, and navigate an intersection, all while obeying given rules of the road.

Robert DiPietro, Christian Rupprecht, Nassir Navab et al.

Recurrent neural networks (RNNs) have achieved state-of-the-art performance on many diverse tasks, from machine translation to surgical activity recognition, yet training RNNs to capture long-term dependencies remains difficult. To date, the vast majority of successful RNN architectures alleviate this problem using nearly-additive connections between states, as introduced by long short-term memory (LSTM). We take an orthogonal approach and introduce MIST RNNs, a NARX RNN architecture that allows direct connections from the very distant past. We show that MIST RNNs 1) exhibit superior vanishing-gradient properties in comparison to LSTM and previously-proposed NARX RNNs; 2) are far more efficient than previously-proposed NARX RNN architectures, requiring even fewer computations than LSTM; and 3) improve performance substantially over LSTM and Clockwork RNNs on tasks requiring very long-term dependencies.