Philippos Mordohai

Weihan Wang, Bharat Joshi, Nathaniel Burgdorfer et al.

This paper addresses real-time dense 3D reconstruction for a resource-constrained Autonomous Underwater Vehicle (AUV). Underwater vision-guided operations are among the most challenging as they combine 3D motion in the presence of external forces, limited visibility, and absence of global positioning. Obstacle avoidance and effective path planning require online dense reconstructions of the environment. Autonomous operation is central to environmental monitoring, marine archaeology, resource utilization, and underwater cave exploration. To address this problem, we propose to use SVIn2, a robust VIO method, together with a real-time 3D reconstruction pipeline. We provide extensive evaluation on four challenging underwater datasets. Our pipeline produces comparable reconstruction with that of COLMAP, the state-of-the-art offline 3D reconstruction method, at high frame rates on a single CPU.

Liyan Chen, Weihan Wang, Philippos Mordohai

We present a new loss function for joint disparity and uncertainty estimation in deep stereo matching. Our work is motivated by the need for precise uncertainty estimates and the observation that multi-task learning often leads to improved performance in all tasks. We show that this can be achieved by requiring the distribution of uncertainty to match the distribution of disparity errors via a KL divergence term in the network's loss function. A differentiable soft-histogramming technique is used to approximate the distributions so that they can be used in the loss. We experimentally assess the effectiveness of our approach and observe significant improvements in both disparity and uncertainty prediction on large datasets.

Weihan Wang, Jiani Li, Yuhang Ming et al.

Visual-inertial initialization can be classified into joint and disjoint approaches. Joint approaches tackle both the visual and the inertial parameters together by aligning observations from feature-bearing points based on IMU integration then use a closed-form solution with visual and acceleration observations to find initial velocity and gravity. In contrast, disjoint approaches independently solve the Structure from Motion (SFM) problem and determine inertial parameters from up-to-scale camera poses obtained from pure monocular SLAM. However, previous disjoint methods have limitations, like assuming negligible acceleration bias impact or accurate rotation estimation by pure monocular SLAM. To address these issues, we propose EDI, a novel approach for fast, accurate, and robust visual-inertial initialization. Our method incorporates an Error-state Kalman Filter (ESKF) to estimate gyroscope bias and correct rotation estimates from monocular SLAM, overcoming dependence on pure monocular SLAM for rotation estimation. To estimate the scale factor without prior information, we offer a closed-form solution for initial velocity, scale, gravity, and acceleration bias estimation. To address gravity and acceleration bias coupling, we introduce weights in the linear least-squares equations, ensuring acceleration bias observability and handling outliers. Extensive evaluation on the EuRoC dataset shows that our method achieves an average scale error of 5.8% in less than 3 seconds, outperforming other state-of-the-art disjoint visual-inertial initialization approaches, even in challenging environments and with artificial noise corruption.

Nathaniel Burgdorfer, Philippos Mordohai

We introduce a learning-based depth map fusion framework that accepts a set of depth and confidence maps generated by a Multi-View Stereo (MVS) algorithm as input and improves them. This is accomplished by integrating volumetric visibility constraints that encode long-range surface relationships across different views into an end-to-end trainable architecture. We also introduce a depth search window estimation sub-network trained jointly with the larger fusion sub-network to reduce the depth hypothesis search space along each ray. Our method learns to model depth consensus and violations of visibility constraints directly from the data; effectively removing the necessity of fine-tuning fusion parameters. Extensive experiments on MVS datasets show substantial improvements in the accuracy of the output fused depth and confidence maps.

Changjiang Cai, Philippos Mordohai

Deep networks for stereo matching typically leverage 2D or 3D convolutional encoder-decoder architectures to aggregate cost and regularize the cost volume for accurate disparity estimation. Due to content-insensitive convolutions and down-sampling and up-sampling operations, these cost aggregation mechanisms do not take full advantage of the information available in the images. Disparity maps suffer from over-smoothing near occlusion boundaries, and erroneous predictions in thin structures. In this paper, we show how deep adaptive filtering and differentiable semi-global aggregation can be integrated in existing 2D and 3D convolutional networks for end-to-end stereo matching, leading to improved accuracy. The improvements are due to utilizing RGB information from the images as a signal to dynamically guide the matching process, in addition to being the signal we attempt to match across the images. We show extensive experimental results on the KITTI 2015 and Virtual KITTI 2 datasets comparing four stereo networks (DispNetC, GCNet, PSMNet and GANet) after integrating four adaptive filters (segmentation-aware bilateral filtering, dynamic filtering networks, pixel adaptive convolution and semi-global aggregation) into their architectures. Our code is available at https://github.com/ccj5351/DAFStereoNets.

Changjiang Cai, Matteo Poggi, Stefano Mattoccia et al.

End-to-end deep networks represent the state of the art for stereo matching. While excelling on images framing environments similar to the training set, major drops in accuracy occur in unseen domains (e.g., when moving from synthetic to real scenes). In this paper we introduce a novel family of architectures, namely Matching-Space Networks (MS-Nets), with improved generalization properties. By replacing learning-based feature extraction from image RGB values with matching functions and confidence measures from conventional wisdom, we move the learning process from the color space to the Matching Space, avoiding over-specialization to domain specific features. Extensive experimental results on four real datasets highlight that our proposal leads to superior generalization to unseen environments over conventional deep architectures, keeping accuracy on the source domain almost unaltered. Our code is available at https://github.com/ccj5351/MS-Nets.

Bo Sun, Hao Kang, Li Guan et al.

We present a deep learning model, dubbed Glissando-Net, to simultaneously estimate the pose and reconstruct the 3D shape of objects at the category level from a single RGB image. Previous works predominantly focused on either estimating poses(often at the instance level), or reconstructing shapes, but not both. Glissando-Net is composed of two auto-encoders that are jointly trained, one for RGB images and the other for point clouds. We embrace two key design choices in Glissando-Net to achieve a more accurate prediction of the 3D shape and pose of the object given a single RGB image as input. First, we augment the feature maps of the point cloud encoder and decoder with transformed feature maps from the image decoder, enabling effective 2D-3D interaction in both training and prediction. Second, we predict both the 3D shape and pose of the object in the decoder stage. This way, we better utilize the information in the 3D point clouds presented only in the training stage to train the network for more accurate prediction. We jointly train the two encoder-decoders for RGB and point cloud data to learn how to pass latent features to the point cloud decoder during inference. In testing, the encoder of the 3D point cloud is discarded. The design of Glissando-Net is inspired by codeSLAM. Unlike codeSLAM, which targets 3D reconstruction of scenes, we focus on pose estimation and shape reconstruction of objects, and directly predict the object pose and a pose invariant 3D reconstruction without the need of the code optimization step. Extensive experiments, involving both ablation studies and comparison with competing methods, demonstrate the efficacy of our proposed method, and compare favorably with the state-of-the-art.

Eric C. Joyce, Qianwen Zhao, Nathaniel Burgdorfer et al.

Deep object pose estimators are notoriously overconfident. A grasping agent that both estimates the 6-DoF pose of a target object and predicts the uncertainty of its own estimate could avoid task failure by choosing not to act under high uncertainty. Even though object pose estimation improves and uncertainty quantification research continues to make strides, few studies have connected them to the downstream task of robotic grasping. We propose a method for training lightweight, deep networks to predict whether a grasp guided by an image-based pose estimate will succeed before that grasp is attempted. We generate training data for our networks via object pose estimation on real images and simulated grasping. We also find that, despite high object variability in grasping trials, networks benefit from training on all objects jointly, suggesting that a diverse variety of objects can nevertheless contribute to the same goal.

Liyan Chen, Huangying Zhan, Hairong Yin et al.

Effective robotic autonomy in unknown environments demands proactive exploration and precise understanding of both geometry and semantics. In this paper, we propose ActiveSGM, an active semantic mapping framework designed to predict the informativeness of potential observations before execution. Built upon a 3D Gaussian Splatting (3DGS) mapping backbone, our approach employs semantic and geometric uncertainty quantification, coupled with a sparse semantic representation, to guide exploration. By enabling robots to strategically select the most beneficial viewpoints, ActiveSGM efficiently enhances mapping completeness, accuracy, and robustness to noisy semantic data, ultimately supporting more adaptive scene exploration. Our experiments on the Replica and Matterport3D datasets highlight the effectiveness of ActiveSGM in active semantic mapping tasks.

Nathaniel Burgdorfer, Philippos Mordohai

We introduce an inference-time scene optimization algorithm utilizing triangle soup, a collection of disconnected translucent triangle primitives, as the representation for the geometry and appearance of a scene. Unlike full-rank Gaussian kernels, triangles are a natural, locally-flat proxy for surfaces that can be connected to achieve highly complex geometry. When coupled with per-vertex Spherical Harmonics (SH), triangles provide a rich visual representation without incurring an expensive increase in primitives. We leverage our new representation to incorporate optimization objectives and enforce spatial regularization directly on the underlying primitives. The main differentiator of our approach is the definition and enforcement of soft connectivity forces between triangles during optimization, encouraging explicit, but soft, surface continuity in 3D. Experiments on representative 3D reconstruction and novel view synthesis datasets show improvements in geometric accuracy compared to current state-of-the-art algorithms without sacrificing visual fidelity.

Tejas Mane, Aylar Bayramova, Kostas Daniilidis et al.

We address the problem of estimating the shape of a person's head, defined as the geometry of the complete head surface, from a video taken with a single moving camera, and determining the alignment of the fitted 3D head for all video frames, irrespective of the person's pose. 3D head reconstructions commonly tend to focus on perfecting the face reconstruction, leaving the scalp to a statistical approximation. Our goal is to reconstruct the head model of each person to enable future mixed reality applications. To do this, we recover a dense 3D reconstruction and camera information via structure-from-motion and multi-view stereo. These are then used in a new two-stage fitting process to recover the 3D head shape by iteratively fitting a 3D morphable model of the head with the dense reconstruction in canonical space and fitting it to each person's head, using both traditional facial landmarks and scalp features extracted from the head's segmentation mask. Our approach recovers consistent geometry for varying head shapes, from videos taken by different people, with different smartphones, and in a variety of environments from living rooms to outdoor spaces.

Matteo Poggi, Fabio Tosi, Konstantinos Batsos et al.

Stereo matching is one of the longest-standing problems in computer vision with close to 40 years of studies and research. Throughout the years the paradigm has shifted from local, pixel-level decision to various forms of discrete and continuous optimization to data-driven, learning-based methods. Recently, the rise of machine learning and the rapid proliferation of deep learning enhanced stereo matching with new exciting trends and applications unthinkable until a few years ago. Interestingly, the relationship between these two worlds is two-way. While machine, and especially deep, learning advanced the state-of-the-art in stereo matching, stereo itself enabled new ground-breaking methodologies such as self-supervised monocular depth estimation based on deep networks. In this paper, we review recent research in the field of learning-based depth estimation from single and binocular images highlighting the synergies, the successes achieved so far and the open challenges the community is going to face in the immediate future.

Liyan Chen, Philippos Mordohai, Sergul Aydore

Dropout as a common regularizer to prevent overfitting in deep neural networks has been less effective in convolutional layers than in fully connected layers. This is because Dropout drops features randomly, without considering local structure. When features are spatially correlated, as in the case of convolutional layers, information from the dropped features can still propagate to subsequent layers via neighboring features. To address this problem, structured forms of Dropout have been proposed. A drawback of these methods is that they do not adapt to the data. In this work, we leverage the structure in the outputs of convolutional layers and introduce a novel structured regularization method named DropCluster. Our approach clusters features in convolutional layers, and drops the resulting clusters randomly during training iterations. Experiments on CIFAR-10/100, SVHN, and APPA-REAL datasets demonstrate that our approach is effective and controls overfitting better than other approaches.

Bo Sun, Philippos Mordohai

Plane detection in 3D point clouds is a crucial pre-processing step for applications such as point cloud segmentation, semantic mapping and SLAM. In contrast to many recent plane detection methods that are only applicable on organized point clouds, our work is targeted to unorganized point clouds that do not permit a 2D parametrization. We compare three methods for detecting planes in point clouds efficiently. One is a novel method proposed in this paper that generates plane hypotheses by sampling from a set of points with estimated normals. We named this method Oriented Point Sampling (OPS) to contrast with more conventional techniques that require the sampling of three unoriented points to generate plane hypotheses. We also implemented an efficient plane detection method based on local sampling of three unoriented points and compared it with OPS and the 3D-KHT algorithm, which is based on octrees, on the detection of planes on 10,000 point clouds from the SUN RGB-D dataset.

Konstantinos Batsos, Changjiang Cai, Philippos Mordohai

Recently, there has been a paradigm shift in stereo matching with learning-based methods achieving the best results on all popular benchmarks. The success of these methods is due to the availability of training data with ground truth; training learning-based systems on these datasets has allowed them to surpass the accuracy of conventional approaches based on heuristics and assumptions. Many of these assumptions, however, had been validated extensively and hold for the majority of possible inputs. In this paper, we generate a matching volume leveraging both data with ground truth and conventional wisdom. We accomplish this by coalescing diverse evidence from a bidirectional matching process via random forest classifiers. We show that the resulting matching volume estimation method achieves similar accuracy to purely data-driven alternatives on benchmarks and that it generalizes to unseen data much better. In fact, the results we submitted to the KITTI and ETH3D benchmarks were generated using a classifier trained on the Middlebury 2014 dataset.

Charles Freundlich, Yan Zhang, Alex Zihao Zhu et al.

In this paper, we address the problem of controlling a mobile stereo camera under image quantization noise. Assuming that a pair of images of a set of targets is available, the camera moves through a sequence of Next-Best-Views (NBVs), i.e., a sequence of views that minimize the trace of the targets' cumulative state covariance, constructed using a realistic model of the stereo rig that captures image quantization noise and a Kalman Filter (KF) that fuses the observation history with new information. The proposed algorithm decomposes control into two stages: first the NBV is computed in the camera relative coordinates, and then the camera moves to realize this view in the fixed global coordinate frame. This decomposition allows the camera to drive to a new pose that effectively realizes the NBV in camera coordinates while satisfying Field-of-View constraints in global coordinates, a task that is particularly challenging using complex sensing models. We provide simulations and real experiments that illustrate the ability of the proposed mobile camera system to accurately localize sets of targets. We also propose a novel data-driven technique to characterize unmodeled uncertainty, such as calibration errors, at the pixel level and show that this method ensures stability of the KF.

Leizer Teran, Philippos Mordohai

The task of detecting the interest points in 3D meshes has typically been handled by geometric methods. These methods, while greatly describing human preference, can be ill-equipped for handling the variety and subjectivity in human responses. Different tasks have different requirements for interest point detection; some tasks may necessitate high precision while other tasks may require high recall. Sometimes points with high curvature may be desirable, while in other cases high curvature may be an indication of noise. Geometric methods lack the required flexibility to adapt to such changes. As a consequence, interest point detection seems to be well suited for machine learning methods that can be trained to match the criteria applied on the annotated training data. In this paper, we formulate interest point detection as a supervised binary classification problem using a random forest as our classifier. Among other challenges, we are faced with an imbalanced learning problem due to the substantial difference in the priors between interest and non-interest points. We address this by re-sampling the training set. We validate the accuracy of our method and compare our results to those of five state of the art methods on a new, standard benchmark.