Nan Yang

Michael Yeung, Leonardo Rundo, Yang Nan et al.

The Dice similarity coefficient (DSC) is both a widely used metric and loss function for biomedical image segmentation due to its robustness to class imbalance. However, it is well known that the DSC loss is poorly calibrated, resulting in overconfident predictions that cannot be usefully interpreted in biomedical and clinical practice. Performance is often the only metric used to evaluate segmentations produced by deep neural networks, and calibration is often neglected. However, calibration is important for translation into biomedical and clinical practice, providing crucial contextual information to model predictions for interpretation by scientists and clinicians. In this study, we provide a simple yet effective extension of the DSC loss, named the DSC++ loss, that selectively modulates the penalty associated with overconfident, incorrect predictions. As a standalone loss function, the DSC++ loss achieves significantly improved calibration over the conventional DSC loss across six well-validated open-source biomedical imaging datasets, including both 2D binary and 3D multi-class segmentation tasks. Similarly, we observe significantly improved calibration when integrating the DSC++ loss into four DSC-based loss functions. Finally, we use softmax thresholding to illustrate that well calibrated outputs enable tailoring of recall-precision bias, which is an important post-processing technique to adapt the model predictions to suit the biomedical or clinical task. The DSC++ loss overcomes the major limitation of the DSC loss, providing a suitable loss function for training deep learning segmentation models for use in biomedical and clinical practice. Source code is available at: https://github.com/mlyg/DicePlusPlus.

Muwei Jian, Hongyu Chen, Zaiyong Zhang et al.

Recently, Computer-Aided Diagnosis (CAD) systems have emerged as indispensable tools in clinical diagnostic workflows, significantly alleviating the burden on radiologists. Nevertheless, despite their integration into clinical settings, CAD systems encounter limitations. Specifically, while CAD systems can achieve high performance in the detection of lung nodules, they face challenges in accurately predicting multiple cancer types. This limitation can be attributed to the scarcity of publicly available datasets annotated with expert-level cancer type information. This research aims to bridge this gap by providing publicly accessible datasets and reliable tools for medical diagnosis, facilitating a finer categorization of different types of lung diseases so as to offer precise treatment recommendations. To achieve this objective, we curated a diverse dataset of lung Computed Tomography (CT) images, comprising 330 annotated nodules (nodules are labeled as bounding boxes) from 95 distinct patients. The quality of the dataset was evaluated using a variety of classical classification and detection models, and these promising results demonstrate that the dataset has a feasible application and further facilitate intelligent auxiliary diagnosis.

Xingxing Zuo, Nan Yang, Nathaniel Merrill et al.

Incrementally recovering 3D dense structures from monocular videos is of paramount importance since it enables various robotics and AR applications. Feature volumes have recently been shown to enable efficient and accurate incremental dense reconstruction without the need to first estimate depth, but they are not able to achieve as high of a resolution as depth-based methods due to the large memory consumption of high-resolution feature volumes. This letter proposes a real-time feature volume-based dense reconstruction method that predicts TSDF (Truncated Signed Distance Function) values from a novel sparsified deep feature volume, which is able to achieve higher resolutions than previous feature volume-based methods, and is favorable in large-scale outdoor scenarios where the majority of voxels are empty. An uncertainty-aware multi-view stereo (MVS) network is leveraged to infer initial voxel locations of the physical surface in a sparse feature volume. Then for refining the recovered 3D geometry, deep features are attentively aggregated from multiview images at potential surface locations, and temporally fused. Besides achieving higher resolutions than before, our method is shown to produce more complete reconstructions with finer detail in many cases. Extensive evaluations on both public and self-collected datasets demonstrate a very competitive real-time reconstruction result for our method compared to state-of-the-art reconstruction methods in both indoor and outdoor settings.

Felix Wimbauer, Nan Yang, Lukas von Stumberg et al.

In this paper, we propose MonoRec, a semi-supervised monocular dense reconstruction architecture that predicts depth maps from a single moving camera in dynamic environments. MonoRec is based on a multi-view stereo setting which encodes the information of multiple consecutive images in a cost volume. To deal with dynamic objects in the scene, we introduce a MaskModule that predicts moving object masks by leveraging the photometric inconsistencies encoded in the cost volumes. Unlike other multi-view stereo methods, MonoRec is able to reconstruct both static and moving objects by leveraging the predicted masks. Furthermore, we present a novel multi-stage training scheme with a semi-supervised loss formulation that does not require LiDAR depth values. We carefully evaluate MonoRec on the KITTI dataset and show that it achieves state-of-the-art performance compared to both multi-view and single-view methods. With the model trained on KITTI, we further demonstrate that MonoRec is able to generalize well to both the Oxford RobotCar dataset and the more challenging TUM-Mono dataset recorded by a handheld camera. Code and related materials will be available at https://vision.in.tum.de/research/monorec.

Lukas von Stumberg, Patrick Wenzel, Nan Yang et al.

We present LM-Reloc -- a novel approach for visual relocalization based on direct image alignment. In contrast to prior works that tackle the problem with a feature-based formulation, the proposed method does not rely on feature matching and RANSAC. Hence, the method can utilize not only corners but any region of the image with gradients. In particular, we propose a loss formulation inspired by the classical Levenberg-Marquardt algorithm to train LM-Net. The learned features significantly improve the robustness of direct image alignment, especially for relocalization across different conditions. To further improve the robustness of LM-Net against large image baselines, we propose a pose estimation network, CorrPoseNet, which regresses the relative pose to bootstrap the direct image alignment. Evaluations on the CARLA and Oxford RobotCar relocalization tracking benchmark show that our approach delivers more accurate results than previous state-of-the-art methods while being comparable in terms of robustness.

L. Koestler, N. Yang, R. Wang et al.

The training of deep-learning-based 3D object detectors requires large datasets with 3D bounding box labels for supervision that have to be generated by hand-labeling. We propose a network architecture and training procedure for learning monocular 3D object detection without 3D bounding box labels. By representing the objects as triangular meshes and employing differentiable shape rendering, we define loss functions based on depth maps, segmentation masks, and ego- and object-motion, which are generated by pre-trained, off-the-shelf networks. We evaluate the proposed algorithm on the real-world KITTI dataset and achieve promising performance in comparison to state-of-the-art methods requiring 3D bounding box labels for training and superior performance to conventional baseline methods.

Patrick Wenzel, Rui Wang, Nan Yang et al.

We present a novel dataset covering seasonal and challenging perceptual conditions for autonomous driving. Among others, it enables research on visual odometry, global place recognition, and map-based re-localization tracking. The data was collected in different scenarios and under a wide variety of weather conditions and illuminations, including day and night. This resulted in more than 350 km of recordings in nine different environments ranging from multi-level parking garage over urban (including tunnels) to countryside and highway. We provide globally consistent reference poses with up-to centimeter accuracy obtained from the fusion of direct stereo visual-inertial odometry with RTK-GNSS. The full dataset is available at https://go.vision.in.tum.de/4seasons.

Nan Yang, Lukas von Stumberg, Rui Wang et al.

We propose D3VO as a novel framework for monocular visual odometry that exploits deep networks on three levels -- deep depth, pose and uncertainty estimation. We first propose a novel self-supervised monocular depth estimation network trained on stereo videos without any external supervision. In particular, it aligns the training image pairs into similar lighting condition with predictive brightness transformation parameters. Besides, we model the photometric uncertainties of pixels on the input images, which improves the depth estimation accuracy and provides a learned weighting function for the photometric residuals in direct (feature-less) visual odometry. Evaluation results show that the proposed network outperforms state-of-the-art self-supervised depth estimation networks. D3VO tightly incorporates the predicted depth, pose and uncertainty into a direct visual odometry method to boost both the front-end tracking as well as the back-end non-linear optimization. We evaluate D3VO in terms of monocular visual odometry on both the KITTI odometry benchmark and the EuRoC MAV dataset.The results show that D3VO outperforms state-of-the-art traditional monocular VO methods by a large margin. It also achieves comparable results to state-of-the-art stereo/LiDAR odometry on KITTI and to the state-of-the-art visual-inertial odometry on EuRoC MAV, while using only a single camera.

Eunah Jung, Nan Yang, Daniel Cremers

We propose the concept of a multi-frame GAN (MFGAN) and demonstrate its potential as an image sequence enhancement for stereo visual odometry in low light conditions. We base our method on an invertible adversarial network to transfer the beneficial features of brightly illuminated scenes to the sequence in poor illumination without costly paired datasets. In order to preserve the coherent geometric cues for the translated sequence, we present a novel network architecture as well as a novel loss term combining temporal and stereo consistencies based on optical flow estimation. We demonstrate that the enhanced sequences improve the performance of state-of-the-art feature-based and direct stereo visual odometry methods on both synthetic and real datasets in challenging illumination. We also show that MFGAN outperforms other state-of-the-art image enhancement and style transfer methods by a large margin in terms of visual odometry.

Rui Wang, Nan Yang, Joerg Stueckler et al.

Scene understanding from images is a challenging problem encountered in autonomous driving. On the object level, while 2D methods have gradually evolved from computing simple bounding boxes to delivering finer grained results like instance segmentations, the 3D family is still dominated by estimating 3D bounding boxes. In this paper, we propose a novel approach to jointly infer the 3D rigid-body poses and shapes of vehicles from a stereo image pair using shape priors. Unlike previous works that geometrically align shapes to point clouds from dense stereo reconstruction, our approach works directly on images by combining a photometric and a silhouette alignment term in the energy function. An adaptive sparse point selection scheme is proposed to efficiently measure the consistency with both terms. In experiments, we show superior performance of our method on 3D pose and shape estimation over the previous geometric approach and demonstrate that our method can also be applied as a refinement step and significantly boost the performances of several state-of-the-art deep learning based 3D object detectors. All related materials and demonstration videos are available at the project page https://vision.in.tum.de/research/vslam/direct-shape.

Nan Yang, Rui Wang, Jörg Stückler et al.

Monocular visual odometry approaches that purely rely on geometric cues are prone to scale drift and require sufficient motion parallax in successive frames for motion estimation and 3D reconstruction. In this paper, we propose to leverage deep monocular depth prediction to overcome limitations of geometry-based monocular visual odometry. To this end, we incorporate deep depth predictions into Direct Sparse Odometry (DSO) as direct virtual stereo measurements. For depth prediction, we design a novel deep network that refines predicted depth from a single image in a two-stage process. We train our network in a semi-supervised way on photoconsistency in stereo images and on consistency with accurate sparse depth reconstructions from Stereo DSO. Our deep predictions excel state-of-the-art approaches for monocular depth on the KITTI benchmark. Moreover, our Deep Virtual Stereo Odometry clearly exceeds previous monocular and deep learning based methods in accuracy. It even achieves comparable performance to the state-of-the-art stereo methods, while only relying on a single camera.

Nan Yang, Rui Wang, Xiang Gao et al.

Monocular visual odometry (VO) and simultaneous localization and mapping (SLAM) have seen tremendous improvements in accuracy, robustness and efficiency, and have gained increasing popularity over recent years. Nevertheless, not so many discussions have been carried out to reveal the influences of three very influential yet easily overlooked aspects: photometric calibration, motion bias and rolling shutter effect. In this work, we evaluate these three aspects quantitatively on the state of the art of direct, feature-based and semi-direct methods, providing the community with useful practical knowledge both for better applying existing methods and developing new algorithms of VO and SLAM. Conclusions (some of which are counter-intuitive) are drawn with both technical and empirical analyses to all of our experiments. Possible improvements on existing methods are directed or proposed, such as a sub-pixel accuracy refinement of ORB-SLAM which boosts its performance.