evin Ornek | Technical University of Munich (original) (raw)

Papers by evin Ornek

Cornell University - arXiv, Nov 26, 2019

Recent advances in artificial intelligence research have led to a profusion of studies that apply... more Recent advances in artificial intelligence research have led to a profusion of studies that apply deep learning to problems in image analysis and natural language processing among others. Additionally, the availability of open-source computational frameworks has lowered the barriers to implementing state-of-the-art methods across multiple domains. Albeit leading to major performance breakthroughs in some tasks, effective dissemination of deep learning algorithms remains challenging, inhibiting reproducibility and benchmarking studies, impeding further validation, and ultimately hindering their effectiveness in the cumulative scientific progress. In developing a platform for sharing research outputs, we present ModelHub.AI (www.modelhub.ai) , a community-driven container-based software engine and platform for the structured dissemination of deep learning models. For contributors, the engine controls data flow throughout the inference cycle, while the contributor-facing standard template exposes model-specific functions including inference, as well as pre-and post-processing. Python and RESTful Application programming interfaces (APIs) enable users to interact with models hosted on ModelHub.AI and allows both researchers and developers to utilize models out-of-the-box. ModelHub.AI is domain-, data-, and framework-agnostic, catering to different workflows and contributors' preferences.

Lecture Notes in Computer Science

Surgical procedures are conducted in highly complex operating rooms (OR), comprising different ac... more Surgical procedures are conducted in highly complex operating rooms (OR), comprising different actors, devices, and interactions. To date, only medically trained human experts are capable of understanding all the links and interactions in such a demanding environment. This paper aims to bring the community one step closer to automated, holistic and semantic understanding and modeling of OR domain. Towards this goal, for the first time, we propose using semantic scene graphs (SSG) to describe and summarize the surgical scene. The nodes of the scene graphs represent different actors and objects in the room, such as medical staff, patients, and medical equipment, whereas edges are the relationships between them. To validate the possibilities of the proposed representation, we create the first publicly available 4D surgical SSG dataset, 4D-OR, containing ten simulated total knee replacement surgeries recorded with six RGB-D sensors in a realistic OR simulation center. 4D-OR includes 6734 frames and is richly annotated with SSGs, human and object poses, and clinical roles. We propose an end-to-end neural network-based SSG generation pipeline, with a rate of success of 0.75 macro F1, indeed being able to infer semantic reasoning in the OR. We further demonstrate the representation power of our scene graphs by using it for the problem of clinical role prediction, where we achieve 0.85 macro F1. The code and dataset will be made available upon acceptance.

ArXiv, 2022

There have been numerous recently proposed methods for monocular depth prediction (MDP) coupled w... more There have been numerous recently proposed methods for monocular depth prediction (MDP) coupled with the equally rapid evolution of benchmarking tools. However, we argue that MDP is currently witnessing benchmark over-fitting and relying on metrics that are only partially helpful to gauge the usefulness of the predictions for 3D applications. This limits the design and development of novel methods that are truly aware of and improving towards estimating the 3D structure of the scene rather than optimizing 2D-based distances. In this work, we aim to bring structural awareness to MDP, an inherently 3D task, by exhibiting the limits of evaluation metrics towards assessing the quality of the 3D geometry. We propose a set of metrics well suited to evaluate the 3D geometry of MDP approaches and a novel indoor benchmark, RIO-D3D, crucial for the proposed evaluation methodology. Our benchmark is based on a real-world dataset featuring high-quality rendered depth maps obtained from RGB-D rec...

ArXiv, 2021

Parts represent a basic unit of geometric and semantic similarity across different objects. We ar... more Parts represent a basic unit of geometric and semantic similarity across different objects. We argue that part knowledge should be composable beyond the observed object classes. Towards this, we present 3D Compositional Zero-shot Learning as a problem of part generalization from seen to unseen object classes for semantic segmentation. We provide a structured study through benchmarking the task with the proposed Compositional-PartNet dataset. This dataset is created by processing the original PartNet to maximize part overlap across different objects. The existing point cloud part segmentation methods fail to generalize to unseen object classes in this setting. As a solution, we propose DeCompositional Consensus, which combines a part segmentation network with a part scoring network. The key intuition to our approach is that a segmentation mask over some parts should have a consensus with its part scores when each part is taken apart. The two networks reason over different part combin...

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018

Reliable and real-time 3D reconstruction and localization functionality is a crucial prerequisite... more Reliable and real-time 3D reconstruction and localization functionality is a crucial prerequisite for the navigation of actively controlled capsule endoscopic robots as an emerging, minimally invasive diagnostic and therapeutic technology for use in the gastrointestinal (GI) tract. In this study, we propose a fully dense, non-rigidly deformable, strictly realtime, intraoperative map fusion approach for actively controlled endoscopic capsule robot applications which combines magnetic and vision-based localization, with non-rigid deformations based frame-to-model map fusion. The performance of the proposed method is demonstrated using four different ex-vivo porcine stomach models. Across different trajectories of varying speed and complexity, and four different endoscopic cameras, the root mean square surface reconstruction errors 1.58 to 2.17 cm.

IEEE Robotics and Automation Letters, 2020

This work proposes a novel SLAM framework for stereo and visual inertial odometry estimation. It ... more This work proposes a novel SLAM framework for stereo and visual inertial odometry estimation. It builds an efficient and robust parametrization of co-planar points and lines which leverages specific geometric constraints to improve camera pose optimization in terms of both efficiency and accuracy. The pipeline consists of extracting 2D points and lines, predicting planar regions and filtering the outliers via RANSAC. Our parametrization scheme then represents co-planar points and lines as their 2D image coordinates and parameters of planes. We demonstrate the effectiveness of the proposed method by comparing it to traditional parametrizations in a novel Monte-Carlo simulation set. Further, the whole stereo SLAM and VIO system is compared with state-of-the-art methods on the public real-world dataset EuRoC. Our method shows better results in terms of accuracy and efficiency than the stateof-the-art. The code is released at https://github.com/LiXin97/ Co-Planar-Parametrization.

In this paper, we examined the zero-shot activity recognition task with the usage of videos. We i... more In this paper, we examined the zero-shot activity recognition task with the usage of videos. We introduce an auto-encoder based model to construct a multimodal joint embedding space between the visual and textual manifolds. On the visual side, we used activity videos and a state-of-the-art 3D convolutional action recognition network to extract the features. On the textual side, we worked with GloVe word embeddings. The zero-shot recognition results are evaluated by top-n accuracy. Then, the manifold learning ability is measured by mean Nearest Neighbor Overlap. In the end, we provide an extensive discussion over the results and the future directions.

From a computer science viewpoint, a surgical domain model needs to be a conceptual one incorpora... more From a computer science viewpoint, a surgical domain model needs to be a conceptual one incorporating both behavior and data. It should therefore model actors, devices, tools, their complex interactions and data flow. To capture and model these, we take advantage of the latest computer vision methodologies for generating 3D scene graphs from camera views. We then introduce the Multimodal Semantic Scene Graph (MSSG) which aims at providing a unified symbolic, spatiotemporal and semantic representation of surgical procedures. This methodology aims at modeling the relationship between different components in surgical domain including medical staff, imaging systems, and surgical devices, opening the path towards holistic understanding and modeling of surgical procedures. We then use MSSG to introduce a dynamically generated graphical user interface tool for surgical procedure analysis which could be used for many applications including process optimization, OR design and automatic repor...

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018

In the last decade, many medical companies and research groups have tried to convert passive caps... more In the last decade, many medical companies and research groups have tried to convert passive capsule endoscopes as an emerging and minimally invasive diagnostic technology into actively steerable endoscopic capsule robots which will provide more intuitive disease detection, targeted drug delivery and biopsy-like operations in the gastrointestinal(GI) tract. In this study, we introduce a fully unsupervised, realtime odometry and depth learner for monocular endoscopic capsule robots. We establish the supervision by warping view sequences and assigning the re-projection minimization to the loss function, which we adopt in multi-view pose estimation and single-view depth estimation network. Detailed quantitative and qualitative analyses of the proposed framework performed on non-rigidly deformable ex-vivo porcine stomach datasets proves the effectiveness of the method in terms of motion estimation and depth recovery.

IEEE Robotics and Automation Letters, 2022

With the advent of deep learning, estimating depth from a single RGB image has recently received ... more With the advent of deep learning, estimating depth from a single RGB image has recently received a lot of attention, being capable of empowering many different applications ranging from path planning for robotics to computational cinematography. Nevertheless, while the depth maps are in their entirety fairly reliable, the estimates around object discontinuities are still far from satisfactory. This can be attributed to the fact that the convolutional operator naturally aggregates features across object discontinuities, resulting in smooth transitions rather than clear boundaries. Therefore, in order to circumvent this issue, we propose a novel convolutional operator which is explicitly tailored to avoid feature aggregation of different object parts. In particular, our method is based on estimating per-part depth values by means of super-pixels. The proposed convolutional operator, which we dub "Instance Convolution", then only considers each object part individually on the basis of the estimated super-pixels. Our evaluation with respect to the NYUv2, iBims and KITTI datasets demonstrate the advantages of Instance Convolutions over the classical convolution at estimating depth around occlusion boundaries, while producing comparable results elsewhere. Our code is available here 1 .

Cornell University - arXiv, Nov 26, 2019

Recent advances in artificial intelligence research have led to a profusion of studies that apply... more Recent advances in artificial intelligence research have led to a profusion of studies that apply deep learning to problems in image analysis and natural language processing among others. Additionally, the availability of open-source computational frameworks has lowered the barriers to implementing state-of-the-art methods across multiple domains. Albeit leading to major performance breakthroughs in some tasks, effective dissemination of deep learning algorithms remains challenging, inhibiting reproducibility and benchmarking studies, impeding further validation, and ultimately hindering their effectiveness in the cumulative scientific progress. In developing a platform for sharing research outputs, we present ModelHub.AI (www.modelhub.ai) , a community-driven container-based software engine and platform for the structured dissemination of deep learning models. For contributors, the engine controls data flow throughout the inference cycle, while the contributor-facing standard template exposes model-specific functions including inference, as well as pre-and post-processing. Python and RESTful Application programming interfaces (APIs) enable users to interact with models hosted on ModelHub.AI and allows both researchers and developers to utilize models out-of-the-box. ModelHub.AI is domain-, data-, and framework-agnostic, catering to different workflows and contributors' preferences.

Lecture Notes in Computer Science

Surgical procedures are conducted in highly complex operating rooms (OR), comprising different ac... more Surgical procedures are conducted in highly complex operating rooms (OR), comprising different actors, devices, and interactions. To date, only medically trained human experts are capable of understanding all the links and interactions in such a demanding environment. This paper aims to bring the community one step closer to automated, holistic and semantic understanding and modeling of OR domain. Towards this goal, for the first time, we propose using semantic scene graphs (SSG) to describe and summarize the surgical scene. The nodes of the scene graphs represent different actors and objects in the room, such as medical staff, patients, and medical equipment, whereas edges are the relationships between them. To validate the possibilities of the proposed representation, we create the first publicly available 4D surgical SSG dataset, 4D-OR, containing ten simulated total knee replacement surgeries recorded with six RGB-D sensors in a realistic OR simulation center. 4D-OR includes 6734 frames and is richly annotated with SSGs, human and object poses, and clinical roles. We propose an end-to-end neural network-based SSG generation pipeline, with a rate of success of 0.75 macro F1, indeed being able to infer semantic reasoning in the OR. We further demonstrate the representation power of our scene graphs by using it for the problem of clinical role prediction, where we achieve 0.85 macro F1. The code and dataset will be made available upon acceptance.

ArXiv, 2022

There have been numerous recently proposed methods for monocular depth prediction (MDP) coupled w... more There have been numerous recently proposed methods for monocular depth prediction (MDP) coupled with the equally rapid evolution of benchmarking tools. However, we argue that MDP is currently witnessing benchmark over-fitting and relying on metrics that are only partially helpful to gauge the usefulness of the predictions for 3D applications. This limits the design and development of novel methods that are truly aware of and improving towards estimating the 3D structure of the scene rather than optimizing 2D-based distances. In this work, we aim to bring structural awareness to MDP, an inherently 3D task, by exhibiting the limits of evaluation metrics towards assessing the quality of the 3D geometry. We propose a set of metrics well suited to evaluate the 3D geometry of MDP approaches and a novel indoor benchmark, RIO-D3D, crucial for the proposed evaluation methodology. Our benchmark is based on a real-world dataset featuring high-quality rendered depth maps obtained from RGB-D rec...

ArXiv, 2021

Parts represent a basic unit of geometric and semantic similarity across different objects. We ar... more Parts represent a basic unit of geometric and semantic similarity across different objects. We argue that part knowledge should be composable beyond the observed object classes. Towards this, we present 3D Compositional Zero-shot Learning as a problem of part generalization from seen to unseen object classes for semantic segmentation. We provide a structured study through benchmarking the task with the proposed Compositional-PartNet dataset. This dataset is created by processing the original PartNet to maximize part overlap across different objects. The existing point cloud part segmentation methods fail to generalize to unseen object classes in this setting. As a solution, we propose DeCompositional Consensus, which combines a part segmentation network with a part scoring network. The key intuition to our approach is that a segmentation mask over some parts should have a consensus with its part scores when each part is taken apart. The two networks reason over different part combin...

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018

Reliable and real-time 3D reconstruction and localization functionality is a crucial prerequisite... more Reliable and real-time 3D reconstruction and localization functionality is a crucial prerequisite for the navigation of actively controlled capsule endoscopic robots as an emerging, minimally invasive diagnostic and therapeutic technology for use in the gastrointestinal (GI) tract. In this study, we propose a fully dense, non-rigidly deformable, strictly realtime, intraoperative map fusion approach for actively controlled endoscopic capsule robot applications which combines magnetic and vision-based localization, with non-rigid deformations based frame-to-model map fusion. The performance of the proposed method is demonstrated using four different ex-vivo porcine stomach models. Across different trajectories of varying speed and complexity, and four different endoscopic cameras, the root mean square surface reconstruction errors 1.58 to 2.17 cm.

IEEE Robotics and Automation Letters, 2020

This work proposes a novel SLAM framework for stereo and visual inertial odometry estimation. It ... more This work proposes a novel SLAM framework for stereo and visual inertial odometry estimation. It builds an efficient and robust parametrization of co-planar points and lines which leverages specific geometric constraints to improve camera pose optimization in terms of both efficiency and accuracy. The pipeline consists of extracting 2D points and lines, predicting planar regions and filtering the outliers via RANSAC. Our parametrization scheme then represents co-planar points and lines as their 2D image coordinates and parameters of planes. We demonstrate the effectiveness of the proposed method by comparing it to traditional parametrizations in a novel Monte-Carlo simulation set. Further, the whole stereo SLAM and VIO system is compared with state-of-the-art methods on the public real-world dataset EuRoC. Our method shows better results in terms of accuracy and efficiency than the stateof-the-art. The code is released at https://github.com/LiXin97/ Co-Planar-Parametrization.

In this paper, we examined the zero-shot activity recognition task with the usage of videos. We i... more In this paper, we examined the zero-shot activity recognition task with the usage of videos. We introduce an auto-encoder based model to construct a multimodal joint embedding space between the visual and textual manifolds. On the visual side, we used activity videos and a state-of-the-art 3D convolutional action recognition network to extract the features. On the textual side, we worked with GloVe word embeddings. The zero-shot recognition results are evaluated by top-n accuracy. Then, the manifold learning ability is measured by mean Nearest Neighbor Overlap. In the end, we provide an extensive discussion over the results and the future directions.

From a computer science viewpoint, a surgical domain model needs to be a conceptual one incorpora... more From a computer science viewpoint, a surgical domain model needs to be a conceptual one incorporating both behavior and data. It should therefore model actors, devices, tools, their complex interactions and data flow. To capture and model these, we take advantage of the latest computer vision methodologies for generating 3D scene graphs from camera views. We then introduce the Multimodal Semantic Scene Graph (MSSG) which aims at providing a unified symbolic, spatiotemporal and semantic representation of surgical procedures. This methodology aims at modeling the relationship between different components in surgical domain including medical staff, imaging systems, and surgical devices, opening the path towards holistic understanding and modeling of surgical procedures. We then use MSSG to introduce a dynamically generated graphical user interface tool for surgical procedure analysis which could be used for many applications including process optimization, OR design and automatic repor...

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018

In the last decade, many medical companies and research groups have tried to convert passive caps... more In the last decade, many medical companies and research groups have tried to convert passive capsule endoscopes as an emerging and minimally invasive diagnostic technology into actively steerable endoscopic capsule robots which will provide more intuitive disease detection, targeted drug delivery and biopsy-like operations in the gastrointestinal(GI) tract. In this study, we introduce a fully unsupervised, realtime odometry and depth learner for monocular endoscopic capsule robots. We establish the supervision by warping view sequences and assigning the re-projection minimization to the loss function, which we adopt in multi-view pose estimation and single-view depth estimation network. Detailed quantitative and qualitative analyses of the proposed framework performed on non-rigidly deformable ex-vivo porcine stomach datasets proves the effectiveness of the method in terms of motion estimation and depth recovery.

IEEE Robotics and Automation Letters, 2022

With the advent of deep learning, estimating depth from a single RGB image has recently received ... more With the advent of deep learning, estimating depth from a single RGB image has recently received a lot of attention, being capable of empowering many different applications ranging from path planning for robotics to computational cinematography. Nevertheless, while the depth maps are in their entirety fairly reliable, the estimates around object discontinuities are still far from satisfactory. This can be attributed to the fact that the convolutional operator naturally aggregates features across object discontinuities, resulting in smooth transitions rather than clear boundaries. Therefore, in order to circumvent this issue, we propose a novel convolutional operator which is explicitly tailored to avoid feature aggregation of different object parts. In particular, our method is based on estimating per-part depth values by means of super-pixels. The proposed convolutional operator, which we dub "Instance Convolution", then only considers each object part individually on the basis of the estimated super-pixels. Our evaluation with respect to the NYUv2, iBims and KITTI datasets demonstrate the advantages of Instance Convolutions over the classical convolution at estimating depth around occlusion boundaries, while producing comparable results elsewhere. Our code is available here 1 .