Camera Motion Estimation Research Articles

RGB-D visual odometry by constructing and matching features at superpixel level

Abstract Visual odometry (VO) is a key technology for estimating camera motion from captured images. In this paper, we propose a novel RGB-D visual odometry by constructing and matching features at the superpixel level that represents better adaptability in different environments than state-of-the-art solutions. Superpixels are content-sensitive and perform well in information aggregation. They could thus characterize the complexity of the environment. Firstly, we designed the superpixel-based feature SegPatch and its corresponding 3D representation MapPatch. By using the neighboring information, SegPatch robustly represents its distinctiveness in various environments with different texture densities. Due to the inclusion of depth measurement, the MapPatch constructs the scene structurally. Then, the distance between SegPatches is defined to characterize the regional similarity. We use the graph search method in scale space for searching and matching. As a result, the accuracy and efficiency of matching process are improved. Additionally, we minimize the reprojection error between the matched SegPatches and estimate camera poses through all these correspondences. Our proposed VO is evaluated on the TUM dataset both quantitatively and qualitatively, showing good balance to adapt to the environment under different realistic conditions.

Unsupervised Monocular Depth Estimation With Channel and Spatial Attention.

Understanding 3-D scene geometry from videos is a fundamental topic in visual perception. In this article, we propose an unsupervised monocular depth and camera motion estimation framework using unlabeled monocular videos to overcome the limitation of acquiring per-pixel ground-truth depth at scale. The photometric loss couples the depth network and pose network together and is essential to the unsupervised method, which is based on warping nearby views to target using the estimated depth and pose. We introduce the channelwise attention mechanism to dig into the relationship between channels and introduce the spatialwise attention mechanism to utilize the inner-spatial relationship of features. Both of them applied in depth networks can better activate the feature information between different convolutional layers and extract more discriminative features. In addition, we apply the Sobel boundary to our edge-aware smoothness for more reasonable accuracy, and clearer boundaries and structures. All of these help to close the gap with fully supervised methods and show high-quality state-of-the-art results on the KITTI benchmark and great generalization performance on the Make3D dataset.

Towards explainable artificial intelligence in deep vision-based odometry

Visual Odometry (VO) is a crucial process for estimating camera motion in real-time based on visual information captured. The emergence of deep learning has significantly transformed VO and Explainable Artificial Intelligence (XAI) in Deep Vision-Based Odometry. This survey paper explores the latest advancements in VO facilitated by deep learning methods, focusing on explainability and interpretability. It provides an overview of state-of-the-art deep learning techniques and dissects each model into its elemental building blocks to understand their explainable and interpretable aspects. The survey also highlights research gaps in optical flow robustness, occlusion and dynamic objects, real-time processing, drift correction, semantic awareness, and sensor integration. The aim is to catalyze future innovations in deep learning-based VO and stimulate dialogue about potential directions for the next wave of research, emphasizing explainability and interpretability as integral components of advanced systems.

Deep Scene Flow Learning: From 2D Images to 3D Point Clouds.

Scene flow describes the 3D motion in a scene. It can be modeled as a single task or as a composite of the auxiliary tasks of depth, camera motion, and optical flow estimation. Deep learning's emergence in recent years has broadened the horizons for new methodologies in estimating these tasks, either as separate tasks or as joint tasks to reconstruct the scene flow. The sequence of images that are either synthesized or captured by a camera is used as input for these methods, which face the challenge of dealing with various situations in images to provide the most accurate motion, such as image quality. Nowadays, images have been superseded by point clouds, which provide 3D information, thereby expediting and enhancing the estimated motion. In this paper, we dig deeply into scene flow estimation in the deep learning era. We provide a comprehensive overview of the important topics regarding both image-based and point-cloud-based methods. In addition, we cover the methodologies for each category, highlighting the network architecture. Furthermore, we provide a comparison between these methods in terms of performance and efficiency. Finally, we conclude this survey with insights and discussions on the open issues and future research directions.

A Temporal Learning Approach to Inpainting Endoscopic Specularities and Its Effect on Image Correspondence

Video streams are utilised to guide minimally-invasive surgery and diagnosis in a wide range of procedures, and many computer-assisted techniques have been developed to automatically analyse them. These approaches can provide additional information to the surgeon such as lesion detection, instrument navigation, or anatomy 3D shape modelling. However, the necessary image features to recognise these patterns are not always reliably detected due to the presence of irregular light patterns such as specular highlight reflections. In this paper, we aim at removing specular highlights from endoscopic videos using machine learning. We propose using a temporal generative adversarial network (GAN) to inpaint the hidden anatomy under specularities, inferring its appearance spatially and from neighbouring frames, where they are not present in the same location. This is achieved using in-vivo data from gastric endoscopy (Hyper Kvasir) in a fully unsupervised manner that relies on the automatic detection of specular highlights. System evaluations show significant improvements to other methods through direct comparison and ablation studies that depict the importance of the network’s temporal and transfer learning components. The generalisability of our system to different surgical setups and procedures was also evaluated qualitatively on in-vivo data of gastric endoscopy and ex-vivo porcine data (SERV-CT, SCARED). We also assess the effect of our method in comparison to other methods on computer vision tasks that underpin 3D reconstruction and camera motion estimation, namely stereo disparity, optical flow, and sparse point feature matching. These are evaluated quantitatively and qualitatively and results show a positive effect of our specular inpainting method on these tasks in a novel comprehensive analysis. Our code and dataset are made available at https://github.com/endomapper/Endo-STTN.

SPSVO: a self-supervised surgical perception stereo visual odometer for endoscopy

AbstractAccurate tracking and reconstruction of surgical scenes is a critical enabling technology toward autonomous robotic surgery. In endoscopic examinations, computer vision has provided assistance in many aspects, such as aiding in diagnosis or scene reconstruction. Estimation of camera motion and scene reconstruction from intra-abdominal images are challenging due to irregular illumination and weak texture of endoscopic images. Current surgical 3D perception algorithms for camera and object pose estimation rely on geometric information (e.g., points, lines, and surfaces) obtained from optical images. Unfortunately, standard hand-crafted local features for pose estimation usually do not perform well in laparoscopic environments. In this paper, a novel self-supervised Surgical Perception Stereo Visual Odometer (SPSVO) framework is proposed to accurately estimate endoscopic pose and better assist surgeons in locating and diagnosing lesions. The proposed SPSVO system combines a self-learning feature extraction method and a self-supervised matching procedure to overcome the adverse effects of irregular illumination in endoscopic images. The framework of the proposed SPSVO includes image pre-processing, feature extraction, stereo matching, feature tracking, keyframe selection, and pose graph optimization. The SPSVO can simultaneously associate the appearance of extracted feature points and textural information for fast and accurate feature tracking. A nonlinear pose graph optimization method is adopted to facilitate the backend process. The effectiveness of the proposed SPSVO framework is demonstrated on a public endoscopic dataset, with the obtained root mean square error of trajectory tracking reaching 0.278 to 0.690 mm. The computation speed of the proposed SPSVO system can reach 71ms per frame.

Rethinking two-dimensional camera motion estimation assessment for digital video stabilization: A camera motion field-based metric

Digital video stabilization aims to remove camera motion jitters through software implementation. The first step of the classical video stabilization methodology is called camera motion estimation, which is usually performed using only RGB frames of the unstable video. Despite recent advances in camera motion estimation strategies, methods classified as two-dimensional are still not properly evaluated, even though it is well known that motion estimation is a crucial step for classical approaches to video stabilization. The main purpose of this work is to draw attention to two-dimensional camera motion estimation assessment and reinforce its importance on video stabilization progress. We proposed a new approach to perform this evaluation using camera motion fields in a pixel-by-pixel comparison and demonstrated through experimental results that our metrics are reliable for diverse scenarios comparing them to image similarity metrics. In addition, we showed and analyzed the results of our metrics for a global and a local method of camera motion estimation. We believe that our assessment and study presented in this work is an important starting point for a more rigorous analysis of this task. In addition, this can be a foundation for coming 2D camera motion estimation methods based on deep learning.

The Right Spin: Learning Object Motion from Rotation-Compensated Flow Fields

A good understanding of geometrical concepts as well as a broad familiarity with objects lead to excellent human perception of moving objects. The human ability to detect and segment moving objects works in the presence of multiple objects, complex background geometry, motion of the observer and even camouflage. How we perceive moving objects so reliably is a longstanding research question in computer vision and borrows findings from related areas such as psychology, cognitive science and physics. One approach to the problem is to teach a deep network to model all of these effects. This is in contrast with the strategy used by human vision, where cognitive processes and body design are tightly coupled and each is responsible for certain aspects of correctly identifying moving objects. Similarly, from the computer vision perspective there is evidence that classical, geometry-based techniques are better suited to the “motion-based” parts of the problem, while deep networks are more suitable for modeling appearance. In this work, we argue that the coupling of camera rotation and camera translation can create complex motion fields that are difficult for a deep network to untangle directly. We present a novel probabilistic model to estimate the camera’s rotation given the motion field. We then rectify the flow field to obtain a rotation-compensated motion field for subsequent segmentation. This strategy of first estimating camera motion, and then allowing a network to learn the remaining parts of the problem, yields improved results on the widely used DAVIS benchmark as well as the more recent motion segmentation data set MoCA (Moving Camouflaged Animals).

PointSLOT: Real-Time Simultaneous Localization and Object Tracking for Dynamic Environment

The applicability of SLAM algorithms is largely limited to the assumption of scene rigidity. Moreover, dynamic object perception is an essential technique for many robotic applications such as autonomous driving and multi-robot collaboration. In this letter, we present PointSLOT, an online, real-time and general stereo system that simultaneously performs SLAM and dynamic object tracking without any artificial priors. Unlike previous SLOT systems, we explicitly identify moving and stationary objects by computing the object motion probabilities, such that features from static objects are utilized to improve the camera pose estimation instead of treating all object regions as outliers. Based on the camera-centric parameterization for the object pose, the trajectories and dynamic landmarks of multi-keyframe objects are efficiently computed through an object-based bundle adjustment approach. The evaluations on KITTI datasets and real-world experiments show that the proposed system achieves comparable results to state-of-the-art solutions on both camera motion estimation and dynamic object tracking. The open-source code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/pkzhou/PointSLOT</uri> .

Wheeled Robot Visual Odometer Based on Two-dimensional Iterative Closest Point Algorithm

According to the two-dimensional motion characteristics of planar motion wheeled robot, the visual odometer was dimensionally reduced in this study. In the feature point matching part of visual odometer, the contour constraint was used to filter out the mismatched feature point pairs (abbreviated as FPP). This method could also filter out the matched FPP, and the feature of FPP was correct color image matches, however, their depth image error was large. This offered higher quality matched FPP for the subsequent interframe motion estimation. Dimension reduction was performed in the interframe motion estimation part, and the two-dimensional Iterative Closest Point (ICP) algorithm was used for camera motion estimation. The experiments indicated that the proposed algorithm effectively improved the computational speed and precision of planar motion wheeled robot visual odometer. This research indicates that the dimension reduction processing of ICP algorithm can effectively improve the operation speed and calculation accuracy of planar motion wheeled robot visual odometry, which provides a good reference and data support for the subsequent research of wheeled robot visual odometry in the future.

A Pose-Only Solution to Visual Reconstruction and Navigation.

Visual navigation and three-dimensional (3D) scene reconstruction are essential for robotics to interact with the surrounding environment. Large-scale scenarios and computational robustness are great challenges facing the research community to achieve this goal. This paper raises a pose-only imaging geometry representation and algorithms that might help solve these challenges. The pose-only representation, equivalent to the classical multiple-view geometry, is discovered to be linearly related to camera global translations, which allows for efficient and robust camera motion estimation. As a result, the spatial feature coordinates can be analytically reconstructed and do not require nonlinear optimization. Comprehensive experiments demonstrate that the computational efficiency of recovering the scene and associated camera poses is significantly improved by 2-4 orders of magnitude.

Enhancing Conventional Geometry-Based Visual Odometry Pipeline Through Integration of Deep Descriptors

Geometry-based Visual Odometry (VO) techniques are renowned in the fields of computer vision and robotics. They use methods from multi-view geometry to estimate camera motion from visual data obtained from one or more cameras. Tracking the camera motion precisely between different views is dependent on the correct estimation of correspondences between salient points of the views. In practice, geometry-based methods are found to be quite effective but do not perform well in challenging cases due to tracking failures caused by abrupt motion, occlusions, textureless and low-light scenes, etc. On the contrary, end-to-end learning from visual data using deep neural networks is an emerging area of research and deals with challenging cases successfully. Despite being computationally expensive, these methods do not outperform their counterparts in conditions favorable to geometry-based methods. Considering these facts in this work, our goal is to integrate deep descriptors to improve the correspondence between image points for tracking in a traditional geometry-based VO pipeline. We propose a simple stereo VO pipeline inspired by popular techniques found in the literature. Two conventional and four deep descriptors have been used in our experiments conducted on various image sequences of the challenging KITTI benchmark dataset. We have determined empirically that deep descriptors can effectively minimize drift in the VO estimates and produce better camera trajectories. The experimental results on the KITTI dataset demonstrate that our VO method performs at par with the state-of-the-art works reported in the literature.

SimVODIS++: Neural Semantic Visual Odometry in Dynamic Environments

Accurate estimation of 3D geometry and camera motion enables a wide range of tasks in robotics and autonomous vehicles. However, the lack of semantics and the performance degradation due to dynamic objects hinder its application to real-world scenarios. To overcome these limitations, we design a novel neural semantic visual odometry (VO) architecture on top of the simultaneous VO, object detection and instance segmentation (SimVODIS) network. Next, we propose an attentive pose estimation architecture with a multi-task learning formulation for handling dynamic objects and VO performance enhancement. The extensive experiments conducted in our work attest that the proposed SimVODIS++ improves the VO performance in dynamic environments. Further, SimVODIS++ focuses on salient regions while excluding feature-less regions. Performing the experiments, we have discovered and fixed the data leakage problem in the conventional experiment setting followed by numerous previous works—which we claim as one of our contributions. We make the source code public.

Single-pass inline pipeline 3D reconstruction using depth camera array

A novel inline inspection (ILI) approach using depth cameras array (DCA) is introduced to create high-fidelity, dense 3D pipeline models. A new camera calibration method is introduced to register the color and the depth information of the cameras into a unified pipe model. By incorporating the calibration outcomes into a robust camera motion estimation approach, dense and complete 3D pipe surface reconstruction is achieved by using only the inline image data collected by a self-powered ILI rover in a single pass through a straight pipeline. The outcomes of the laboratory experiments demonstrate one-millimeter geometrical accuracy and 0.1-pixel photometric accuracy. In the reconstructed model of a longer pipeline, the proposed method generates the dense 3D surface reconstruction model at the millimeter level accuracy with less than 0.5% distance error. The achieved performance highlights its potential as a useful tool for efficient in-line, non-destructive evaluation of pipeline assets.

Deep homography estimation in dynamic surgical scenes for laparoscopic camera motion extraction

ABSTRACT Current laparoscopic camera motion automation relies on rule-based approaches or only focuses on surgical tools. Imitation Learning (IL) methods could alleviate these shortcomings, but have so far been applied to oversimplified setups. Instead of extracting actions from oversimplified setups, in this work we introduce a method that allows to extract a laparoscope holder’s actions from videos of laparoscopic interventions. We synthetically add camera motion to a newly acquired dataset of camera motion free da Vinci surgery image sequences through a novel homography generation algorithm. The synthetic camera motion serves as a supervisory signal for camera motion estimation that is invariant to object and tool motion. We perform an extensive evaluation of state-of-the-art (SOTA) Deep Neural Networks (DNNs) across multiple compute regimes, finding our method transfers from our camera motion free da Vinci surgery dataset to videos of laparoscopic interventions, outperforming classical homography estimation approaches in both, precision by , and runtime on a CPU by .

Accuracy and Speed Improvement of Event Camera Motion Estimation Using a Bird's-Eye View Transformation.

Event cameras are bio-inspired sensors that have a high dynamic range and temporal resolution. This property enables motion estimation from textures with repeating patterns, which is difficult to achieve with RGB cameras. Therefore, motion estimation of an event camera is expected to be applied to vehicle position estimation. An existing method, called contrast maximization, is one of the methods that can be used for event camera motion estimation by capturing road surfaces. However, contrast maximization tends to fall into a local solution when estimating three-dimensional motion, which makes correct estimation difficult. To solve this problem, we propose a method for motion estimation by optimizing contrast in the bird’s-eye view space. Instead of performing three-dimensional motion estimation, we reduced the dimensionality to two-dimensional motion estimation by transforming the event data to a bird’s-eye view using homography calculated from the event camera position. This transformation mitigates the problem of the loss function becoming non-convex, which occurs in conventional methods. As a quantitative experiment, we created event data by using a car simulator and evaluated our motion estimation method, showing an improvement in accuracy and speed. In addition, we conducted estimation from real event data and evaluated the results qualitatively, showing an improvement in accuracy.

Camera Motion Estimation from Image Sequence in Pipe and Construction of 3D Point Cloud of Pipe

Camera Motion Estimation from Image Sequence in Pipe and Construction of 3D Point Cloud of Pipe

IMU-Assisted Online Video Background Identification.

Distinguishing between dynamic foreground objects and a mostly static background is a fundamental problem in many computer vision and computer graphics tasks. This paper presents a novel online video background identification method with the assistance of inertial measurement unit (IMU). Based on the fact that the background motion of a video essentially reflects the 3D camera motion, we leverage IMU data to realize a robust camera motion estimation for identifying background feature points by only investigating a few historical frames. We observe that the displacement of the 2D projection of a scene point caused by camera rotation is depth-invariant, and the rotation estimation by using IMU data can be quite accurate. We thus propose to analyze 2D feature points by decomposing the 2D motion into two components: rotation projection and translation projection. In our method, after establishing the 3D camera rotations, we generate the depth-relevant 2D feature point movement induced by the camera 3D translation. Then, by examining the disparity between inter-frame offset and the projection of estimated 3D camera motion, we can identify the background feature points. In the experiments, our online method is able to run at 30FPS with only 1 frame latency and outperforms state-of-the-art background identification and other relevant methods. Our method directly leads to a better camera motion estimation, which is beneficial to many applications like online video stabilization, SLAM, image stitching, etc.

3D ego-Motion Estimation Using low-Cost mmWave Radars via Radar Velocity Factor for Pose-Graph SLAM

Achieving general 3D motion estimation for all-visibility has been a key challenge in robotics, especially in extreme environments. The widely adopted camera and LiDAR-based motion estimation critically deteriorate under fog or smoke. In this letter, we devised a unique sensor system for 3D velocity estimation by compositing two orthogonal radar sensors. The proposed configuration allowed securing returns from the static objects on the ground. This work aimed at a realistic sensor deployment in a harsh environment by casing the bare sensor rig into a plastic box. As will be shown, the proposed velocity-based ego-motion estimation presented reliable performance over existing point matching-based methods, which degraded when measurement attenuates due to the casing. Furthermore, we introduce a novel radar instant velocity factor for pose-graph simultaneous localization and mapping (SLAM) framework and solve for 3D ego-motion in the integration with IMU. The validation reveals that the proposed method can be applied to estimate general 3D motion in both indoor and outdoor, targetting various visibility and the structureness in the environment.

Depth-based branching level estimation for bronchoscopic navigation

Bronchoscopists rely on navigation systems during bronchoscopy to reduce the risk of getting lost in the complex bronchial tree-like structure and the homogeneous bronchus lumens. We propose a patient-specific branching level estimation method for bronchoscopic navigation because it is vital to identify the branches being examined in the bronchus tree during examination. We estimate the branching level by integrating the changes in the number of bronchial orifices and the camera motions among the frames. We extract the bronchial orifice regions from a depth image, which is generated using a cycle generative adversarial network (CycleGAN) from real bronchoscopic images. We calculate the number of orifice regions using the vertical and horizontal projection profiles of the depth images and obtain the camera-moving direction using the feature point-based camera motion estimation. The changes in the number of bronchial orifices are combined with the camera-moving direction to estimate the branching level. We used three in vivo and one phantom case to train the CycleGAN model and four in vivo cases to validate the proposed method. We manually created the ground truth of the branching level. The experimental results showed that the proposed method can estimate the branching level with an average accuracy of 87.6%. The processing time per frame was about 61 ms. Experimental results show that it is feasible to estimate the branching level using the number of bronchial orifices and camera-motion estimation from real bronchoscopic images.