We present FoundationSLAM, a learning-based monocular dense SLAM system that addresses the absence of geometric consistency in previous flow-based approaches for accurate and robust tracking and mapping. Our core idea is to bridge flow estimation with geometric reasoning by leveraging the guidance from foundation depth models. To this end, we first develop a Hybrid Flow Network that produces geometry-aware correspondences, enabling consistent depth and pose inference across diverse keyframes. To enforce global consistency, we propose a Bi-Consistent Bundle Adjustment Layer that jointly optimizes keyframe pose and depth under multi-view constraints. Furthermore, we introduce a Reliability-Aware Refinement mechanism that dynamically adapts the flow update process by distinguishing between reliable and uncertain regions, forming a closed feedback loop between matching and optimization. Extensive experiments demonstrate that FoundationSLAM achieves superior trajectory accuracy and dense reconstruction quality across multiple challenging datasets, while running in real-time at 18 FPS, demonstrating strong generalization to various scenarios and practical applicability of our method.

Over the past two decades, research in the field of Simultaneous Localization and Mapping (SLAM) has undergone a significant evolution, highlighting its critical role in enabling autonomous exploration of unknown environments. This evo-lution ranges from hand-crafted methods, through the era of deep learning, to more recent developments focused on Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (3DGS) rep-resentations. Recognizing the growing body of research and the absence of a comprehensive survey on the topic, this paper aims to provide the first comprehensive overview of SLAM progress through the lens of the latest advancements in radiance fields. It sheds light on the background, evolutionary path, inherent strengths and limitations, and serves as a fundamental reference to highlight the dynamic progress and specific challenges.

We introduce Stereo Anywhere, a novel stereo-matching framework that combines geometric constraints with robust priors from monocular depth Vision Foundation Models (VFMs). By elegantly coupling these complementary worlds through a dual-branch architecture, we seamlessly integrate stereo matching with learned contextual cues. Following this design, our framework introduces novel cost volume fusion mechanisms that effectively handle critical challenges such as textureless regions, occlusions, and nonLambertian surfaces. Through our novel optical illusion dataset, MonoTrap, and extensive evaluation across multiple benchmarks, we demonstrate that our synthetic-only trained model achieves state-of-the-art results in zero-shot generalization, significantly outperforming existing solutions while showing remarkable robustness to challenging cases such as mirrors and transparencies.

Event cameras capture sparse, high-temporal-resolution visual information, making them particularly suitable for challenging environments with high-speed motion and strongly varying lighting conditions. However, the lack of large datasets with dense ground-truth depth annotations hinders learning-based monocular depth estimation from event data. To address this limitation, we propose a cross-modal distillation paradigm to generate dense proxy labels leveraging a Vision Foundation Model (VFM). Our strategy requires an event stream spatially aligned with RGB frames, a simple setup even available off-the-shelf, and exploits the robustness of large-scale VFMs. Additionally, we propose to adapt VFMs, either a vanilla one like Depth Anything v2 (DAv2), or deriving from it a novel recurrent architecture to infer depth from monocular event cameras. We evaluate our approach with synthetic and real-world datasets, demonstrating that i) our cross-modal paradigm achieves competitive performance compared to fully supervised methods without requiring expensive depth annotations, and ii) our VFM-based models achieve state-of-the-art performance.

NeRF-based SLAM has recently achieved promising results in tracking and reconstruction. However, existing methods face challenges in providing sufficient scene representation, capturing structural information, and maintaining global consistency in scenes emerging significant movement or being forgotten. To this end, we present HS-SLAM to tackle these problems. To enhance scene representation capacity, we propose a hybrid encoding network that combines the complementary strengths of hash-grid, tri-planes, and one-blob, improving the completeness and smoothness of reconstruction. Additionally, we introduce structural supervision by sampling patches of non-local pixels rather than individual rays to better capture the scene structure. To ensure global consistency, we implement an active global bundle adjustment (BA) to eliminate camera drifts and mitigate accumulative errors. Experimental results demonstrate that HS-SLAM outperforms the baselines in tracking and reconstruction accuracy while maintaining the efficiency required for robotics.

Self-supervised single-view depth estimation, trained on video sequences, faces significant challenges when dynamic objects are present in the training data, as they violate the basic multi-view geometry assumptions used to compute photometric losses. We propose a novel approach that leverages the relationship between the depth of moving objects and their ground contact points. By iteratively propagating ground features to moving targets in perceptual layers, we recalibrate the depth of dynamic entities while preserving details. Our method maintains the end-to-end training paradigm without additional networks or complex training procedures. Our experiments demonstrate that our method achieves state-of-the-art performance when estimating depth for dynamic objects and attains superior generalization compared to existing approaches. The relevant experimental code can be accessed at: https://github.com/LiHuanLi/GroundMono

We present Eve3D, a novel framework for dense 3D reconstruction based on 3D Gaussian Splatting (3DGS). While most existing methods rely on imperfect priors derived from pre-trained vision models, Eve3D fully leverages these priors by jointly optimizing both them and the 3DGS backbone. This joint optimization creates a mutually reinforcing cycle: the priors enhance the quality of 3DGS, which in turn refines the priors, further improving the reconstruction. Additionally, Eve3D introduces a novel optimization step based on bundle adjustment, overcoming the limitations of the highly local supervision in standard 3DGS pipelines. Eve3D achieves state-of-the-art results in surface reconstruction and novel view synthesis on the Tanks & Temples, DTU, and Mip-NeRF360 datasets. while retaining fast convergence, highlighting an unprecedented trade-off between accuracy and speed.

Accurate in-cabin depth estimation is critical for advancing automotive safety and occupant comfort. However, existing datasets for in-vehicle scene understanding tasks often fall short in providing sufficient information and scale needed to evaluate existing depth estimation methods. In this paper, we present a novel benchmark tailored for monocular depth estimation in vehicle interiors, containing both near-infrared (NIR) images and corresponding ground truth depth data. Featuring over 41,000 frames captured across 36 distinct vehicles and 45 different passengers, it offers an unprecedented level of variability for this application domain. Evaluation on our testbench of cutting-edge single-view depth models in different flavors, including zero-shot affine-invariant depth estimation or indomain specialization, reveals that current depth estimation approaches, while promising, still have a significant performance gap to overcome before achieving the reliability required for downstream safety-critical applications. In light of its diverse range and complex scenarios, we believe this benchmark could serve as a common reference for further research concerning in-cabin monocular depth estimation.

Abstract Stereo matching is close to hitting a half-century of history, yet witnessed a rapid evolution in the last decade thanks to deep learning. While previous surveys in the late 2010s covered the first stage of this revolution, the last five years of research brought further ground-breaking advancements to the field. This paper aims to fill this gap in a two-fold manner: first, we offer an in-depth examination of the latest developments in deep stereo matching, focusing on the pioneering architectural designs and groundbreaking paradigms that have redefined the field in the 2020s; second, we present a thorough analysis of the critical challenges that have emerged alongside these advances, providing a comprehensive taxonomy of these issues and exploring the state-of-the-art techniques proposed to address them. By reviewing both the architectural innovations and the key challenges, we offer a holistic view of deep stereo matching and highlight the specific areas that require further investigation. To accompany this survey, we maintain a regularly updated project page that catalogs papers on deep stereo matching in our Awesome-Deep-Stereo-Matching repository.

This paper proposes a new framework for depth completion robust against domain-shifting issues. It exploits the generalization capability of modern stereo networks to face depth completion, by processing fictitious stereo pairs obtained through a virtual pattern projection paradigm. Any stereo network or traditional stereo matcher can be seamlessly plugged into our framework, allowing for the deployment of a virtual stereo setup that is future-proof against advancement in the stereo field. Exhaustive experiments on cross-domain generalization support our claims. Hence, we argue that our framework can help depth completion to reach new deployment scenarios.

We introduce a novel approach for adapting deep stereo networks in a collaborative manner. By building over principles of federated learning, we develop a distributed framework allowing for demanding the optimization process to a number of clients deployed in different environments. This makes it possible, for a deep stereo network running on resourced-constrained devices, to capitalize on the adaptation process carried out by other instances of the same architecture, and thus improve its accuracy in challenging environments even when it cannot carry out adaptation on its own. Experimental results show how federated adaptation performs equivalently to on-device adaptation, and even better when dealing with challenging environments.

Estimating depth from images nowadays yields outstanding results, both in terms of in-domain accuracy and generalization. However, we identify two main challenges that remain open in this field: dealing with non-Lambertian materials and effectively processing high-resolution images. Purposely, we propose a novel dataset that includes accurate and dense ground-truth labels at high resolution, featuring scenes containing several specular and transparent surfaces. Our acquisition pipeline leverages a novel deep space-time stereo framework, enabling easy and accurate labeling with sub-pixel precision. The dataset is composed of 606 samples collected in 85 different scenes, each sample includes both a high-resolution pair (12 Mpx) as well as an unbalanced stereo pair (Left: 12 Mpx, Right: 1.1 Mpx), typical of modern mobile devices that mount sensors with different resolutions. Additionally, we provide manually annotated material segmentation masks and 15 K unlabeled samples. The dataset is composed of a train set and two test sets, the latter devoted to the evaluation of stereo and monocular depth estimation networks. Our experiments highlight the open challenges and future research directions in this field.

We propose a framework that combines traditional, hand-crafted algorithms and recent advances in deep learning to obtain high-quality, high-resolution disparity maps from stereo images. By casting the refinement process as a continuous feature sampling strategy, our neural disparity refinement network can estimate an enhanced disparity map at any output resolution. Our solution can process any disparity map produced by classical stereo algorithms, as well as those predicted by modern stereo networks or even different depth-from-images approaches, such as the COLMAP structure-from-motion pipeline. Nonetheless, when deployed in the former configuration, our framework performs at its best in terms of zero-shot generalization from synthetic to real images. Moreover, its continuous formulation allows for easily handling the unbalanced stereo setup very diffused in mobile phones.

We introduce a novel framework for training deep stereo networks effortlessly and without any ground-truth. By leveraging state-of-the-art neural rendering solutions, we generate stereo training data from image sequences collected with a single handheld camera. On top of them, a NeRF-supervised training procedure is carried out, from which we exploit rendered stereo triplets to compensate for occlusions and depth maps as proxy labels. This results in stereo networks capable of predicting sharp and detailed disparity maps. Experimental results show that models trained under this regime yield a 30-40% improvement over existing self-supervised methods on the challenging Middle-bury dataset, filling the gap to supervised models and, most times, outperforming them at zero-shot generalization.

This paper proposes a novel framework integrating the principles of active stereo in standard passive camera systems without a physical pattern projector. We virtually project a pattern over the left and right images according to the sparse measurements obtained from a depth sensor. Any such devices can be seamlessly plugged into our framework, allowing for the deployment of a virtual active stereo setup in any possible environment, overcoming the limitation of pattern projectors, such as limited working range or environmental conditions. Experiments on indoor/outdoor datasets, featuring both long and close-range, support the seamless effectiveness of our approach, boosting the accuracy of both stereo algorithms and deep networks.

Neural implicit representations have recently demonstrated compelling results on dense Simultaneous Localization And Mapping (SLAM) but suffer from the accumulation of errors in camera tracking and distortion in the reconstruction. Purposely, we present GO-SLAM, a deep-learning-based dense visual SLAM framework globally optimizing poses and 3D reconstruction in real-time. Robust pose estimation is at its core, supported by efficient loop closing and online full bundle adjustment, which optimize per frame by utilizing the learned global geometry of the complete history of input frames. Simultaneously, we update the implicit and continuous surface representation on-the-fly to ensure global consistency of 3D reconstruction. Results on various synthetic and real-world datasets demonstrate that GO-SLAM outperforms state-of-the-art approaches at tracking robustness and reconstruction accuracy. Furthermore, GO-SLAM is versatile and can run with monocular, stereo, and RGB-D input.

Inferring the depth of transparent or mirror (ToM) surfaces represents a hard challenge for either sensors, algorithms, or deep networks. We propose a simple pipeline for learning to estimate depth properly for such surfaces with neural networks, without requiring any ground-truth annotation. We unveil how to obtain reliable pseudo labels by in-painting ToM objects in images and processing them with a monocular depth estimation model. These labels can be used to fine-tune existing monocular or stereo networks, to let them learn how to deal with ToM surfaces. Experimental results on the Booster dataset show the dramatic improvements enabled by our remarkably simple proposal.

This paper tackles the challenges of self-supervised monocular depth estimation in indoor scenes caused by large rotation between frames and low texture. We ease the learning process by obtaining coarse camera poses from monocular sequences through multi-view geometry to deal with the former. However, we found that limited by the scale ambiguity across different scenes in the training dataset, a naïve introduction of geometric coarse poses cannot play a positive role in performance improvement, which is counter-intuitive. To address this problem, we propose to refine those poses during training through rotation and translation/scale optimization. To soften the effect of the low texture, we combine the global reasoning of vision transformers with an overfitting-aware, iterative self-distillation mechanism, providing more accurate depth guidance coming from the network itself. Experiments on NYUv2, ScanNet, 7scenes, and KITTI datasets support the effectiveness of each component in our framework, which sets a new state-of-the-art for indoor self-supervised monocular depth estimation, as well as outstanding generalization ability. Code and models are available at https://github.com/zxcqlf/GasMono

We propose X-NeRF, a novel method to learn a Cross-Spectral scene representation given images captured from cameras with different light spectrum sensitivity, based on the Neural Radiance Fields formulation. X-NeRF optimizes camera poses across spectra during training and exploits Normalized Cross-Device Coordinates (NXDC) to render images of different modalities from arbitrary viewpoints, which are aligned and at the same resolution. Experiments on 16 forward-facing scenes, featuring color, multi-spectral and infrared images, confirm the effectiveness of X-NeRF at modeling Cross-Spectral scene representations.

Self-supervised monocular depth estimation is an attractive solution that does not require hard-to-source depth la-bels for training. Convolutional neural networks (CNNs) have recently achieved great success in this task. However, their limited receptive field constrains existing network architectures to reason only locally, dampening the effectiveness of the self-supervised paradigm. In the light of the recent successes achieved by Vision Transformers (ViTs), we propose MonoViT, a brand-new framework combining the global reasoning enabled by ViT models with the flexibility of self-supervised monocular depth estimation. By combining plain convolutions with Transformer blocks, our model can reason locally and globally, yielding depth prediction at a higher level of detail and accuracy, allowing MonoViT to achieve state-of-the-art performance on the established KITTI dataset. Moreover, MonoViT proves its superior generalization capacities on other datasets such as Make3D and DrivingStereo. Source code available at https://github.com/zxcqlf/MonoViT

We present a novel high-resolution and challenging stereo dataset framing indoor scenes annotated with dense and accurate ground-truth disparities. Peculiar to our dataset is the presence of several specular and transparent surfaces, i.e. the main causes of failures for state-of-the-art stereo networks. Our acquisition pipeline leverages a novel deep space-time stereo framework which allows for easy and accurate labeling with sub-pixel precision. We re-lease a total of 419 samples collected in 64 different scenes and annotated with dense ground-truth disparities. Each sample include a high-resolution pair (12 Mpx) as well as an unbalanced pair (Left: 12 Mpx, Right: 1.1 Mpx). Additionally, we provide manually annotated material segmentation masks and 15K unlabeled samples. We evaluate state-of-the-art deep networks based on our dataset, highlighting their limitations in addressing the open challenges in stereo and drawing hints for future research.

We address the problem of registering synchronized color (RGB) and multi-spectral (MS) images featuring very different resolution by solving stereo matching correspondences. Purposely, we introduce a novel RGB-MS dataset framing 13 different scenes in indoor environments and providing a total of 34 image pairs annotated with semi-dense, high-resolution ground-truth labels in the form of disparity maps. To tackle the task, we propose a deep learning architecture trained in a self-supervised manner by exploiting a further RGB camera, required only during training data acquisition. In this setup, we can conveniently learn cross-modal matching in the absence of ground-truth labels by distilling knowledge from an easier RGB-RGB matching task based on a collection of about 11K unlabeled image triplets. Experiments show that the proposed pipeline sets a good performance bar (1.16 pixels average registration error) for future research on this novel, challenging task.

The recent advancements in deep learning have demonstrated that inferring high-quality depth maps from a single image has become feasible and accurate, thanks to convolutional neural networks (CNNs), but how to process such compute- and memory-intensive models on portable and low-power devices remains a concern. Dynamic energy-quality scaling is an interesting yet less explored option in this field. It can improve efficiency through opportunistic computing policies where performances are boosted only when needed, achieving on average substantial energy savings. Implementing such a computing paradigm encompasses the availability of a scalable inference model, which is the target of this work. Specifically, we describe and characterize the design of an energy-quality scalable pyramidal network (EQPyD-Net), a lightweight CNN capable of modulating at runtime the computational effort with minimal memory resources. We describe the architecture of the network and the optimization flow, covering the important aspects that enable the dynamic scaling, namely, the optimized training procedures, the compression stage via fixed-point quantization, and the code optimization for the deployment on commercial low-power CPUs adopted in the edge segment. To assess the effect of the proposed design knobs, we evaluated the prediction quality on the standard KITTI data set and the energy and memory resources on the ARM Cortex-A53 CPU. The collected results demonstrate the flexibility of the proposed network and its energy efficiency. EQPyD-Net can be shifted across five operating points, ranging from a maximum accuracy of 82.2% with 0.4 Frame/J and up to 92.6% of energy savings with 6.1% of accuracy loss, still keeping a compact memory footprint of 5.2 MB for the weights and 38.3 MB (in the worst case) for the processing.

Stereo matching is one of the most popular techniques to estimate dense depth maps by finding the disparity between matching pixels on two, synchronized and rectified images. Alongside with the development of more accurate algorithms, the research community focused on finding good strategies to estimate the reliability, i.e., the confidence, of estimated disparity maps. This information proves to be a powerful cue to naively find wrong matches as well as to improve the overall effectiveness of a variety of stereo algorithms according to different strategies. In this paper, we review more than ten years of developments in the field of confidence estimation for stereo matching. We extensively discuss and evaluate existing confidence measures and their variants, from hand-crafted ones to the most recent, state-of-the-art learning based methods. We study the different behaviors of each measure when applied to a pool of different stereo algorithms and, for the first time in literature, when paired with a state-of-the-art deep stereo network. Our experiments, carried out on five different standard datasets, provide a comprehensive overview of the field, highlighting in particular both strengths and limitations of learning-based strategies.

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.

Depth estimation is crucial in several computer vision applications, and a recent trend in this field aims at inferring such a cue from a single camera. Unfortunately, despite the compelling results achieved, state-of-the-art monocular depth estimation methods are computationally demanding, thus precluding their practical deployment in several application contexts characterized by low-power constraints. Therefore, in this paper, we propose a lightweight Convolutional Neural Network based on a shallow pyramidal architecture, referred to as$\mu $PyD-Net, enabling monocular depth estimation on microcontrollers. The network is trained in a peculiar self-supervised manner leveraging proxy labels obtained through a traditional stereo algorithm. Moreover, we propose optimization strategies aimed at performing computations with quantized 8-bit data and map the high-level description of the network to low-level layers optimized for the target microcontroller architecture. Exhaustive experimental results on standard datasets and an in-depth evaluation with a device belonging to the popular Arm Cortex-M family confirm that obtaining sufficiently accurate monocular depth estimation on microcontrollers is feasible. To the best of our knowledge, our proposal is the first one enabling such remarkable achievement, paving the way for the deployment of monocular depth cues onto the tiny end-nodes of distributed sensor networks.

Single-image depth estimation represents a longstanding challenge in computer vision and although it is an ill-posed problem, deep learning enabled astonishing results leveraging both supervised and self-supervised training paradigms. State-of-the-art solutions achieve remarkably accurate depth estimation from a single image deploying huge deep architectures, requiring powerful dedicated hardware to run in a reasonable amount of time. This overly demanding complexity makes them unsuited for a broad category of applications requiring devices with constrained resources or memory consumption. To tackle this issue, in this paper a family of compact, yet effective CNNs for monocular depth estimation is proposed, by leveraging self-supervision from a binocular stereo rig. Our lightweight architectures, namely PyD-Net and PyD-Net2, compared to complex state-of-the-art trade a small drop in accuracy to drastically reduce the runtime and memory requirements by a factor ranging from$2\times $to$100\times $. Moreover, our networks can run real-time monocular depth estimation on a broad set of embedded or consumer devices, even not equipped with a GPU, by early stopping the inference with negligible (or no) loss in accuracy, making it ideally suited for real applications with strict constraints on hardware resources or power consumption.

We introduce a novel architecture for neural disparity refinement aimed at facilitating deployment of 3D computer vision on cheap and widespread consumer devices, such as mobile phones. Our approach relies on a continuous formulation that enables to estimate a refined disparity map at any arbitrary output resolution. Thereby, it can handle effectively the unbalanced camera setup typical of nowadays mobile phones, which feature both high and low resolution RGB sensors within the same device. Moreover, our neural network can process seamlessly the output of a variety of stereo methods and, by refining the disparity maps computed by a traditional matching algorithm like SGM, it can achieve unpaired zero-shot generalization performance compared to state-of-the-art end-to-end stereo models.

Despite stereo matching accuracy has greatly improved by deep learning in the last few years, recovering sharp boundaries and high-resolution outputs efficiently remains challenging. In this paper, we propose Stereo Mixture Density Networks (SMD-Nets), a simple yet effective learning framework compatible with a wide class of 2D and 3D architectures which ameliorates both issues. Specifically, we exploit bimodal mixture densities as output representation and show that this allows for sharp and precise disparity estimates near discontinuities while explicitly modeling the aleatoric uncertainty inherent in the observations. Moreover, we formulate disparity estimation as a continuous problem in the image domain, allowing our model to query disparities at arbitrary spatial precision. We carry out comprehensive experiments on a new high-resolution and highly realistic synthetic stereo dataset, consisting of stereo pairs at 8Mpx resolution, as well as on real-world stereo datasets. Our experiments demonstrate increased depth accuracy near object boundaries and prediction of ultra high-resolution disparity maps on standard GPUs. We demonstrate the flexibility of our technique by improving the performance of a variety of stereo backbones.

Scene flow is a challenging task aimed at jointly estimating the 3D structure and motion of the sensed environment. Although deep learning solutions achieve outstanding performance in terms of accuracy, these approaches divide the whole problem into standalone tasks (stereo and optical flow) addressing them with independent networks. Such a strategy dramatically increases the complexity of the training procedure and requires power-hungry GPUs to infer scene flow barely at 1 FPS. Conversely, we propose DWARF, a novel and lightweight architecture able to infer full scene flow jointly reasoning about depth and optical flow easily and elegantly trainable end-to-end from scratch. Moreover, since ground truth images for full scene flow are scarce, we propose to leverage on the knowledge learned by networks specialized in stereo or flow, for which much more data are available, to distill proxy annotations. Exhaustive experiments show that i) DWARF runs at about 10 FPS on a single high-end GPU and about 1 FPS on NVIDIA Jetson TX2 embedded at KITTI resolution, with moderate drop in accuracy compared to 10× deeper models, ii) learning from many distilled samples is more effective than from the few, annotated ones available.

Self-supervised paradigms for monocular depth estimation are very appealing since they do not require ground truth annotations at all. Despite the astonishing results yielded by such methodologies, learning to reason about the uncertainty of the estimated depth maps is of paramount importance for practical applications, yet uncharted in the literature. Purposely, we explore for the first time how to estimate the uncertainty for this task and how this affects depth accuracy, proposing a novel peculiar technique specifically designed for self-supervised approaches. On the standard KITTI dataset, we exhaustively assess the performance of each method with different self-supervised paradigms. Such evaluation highlights that our proposal i) always improves depth accuracy significantly and ii) yields state-of-the-art results concerning uncertainty estimation when training on sequences and competitive results uniquely deploying stereo pairs.

Whole understanding of the surroundings is paramount to autonomous systems. Recent works have shown that deep neural networks can learn geometry (depth) and motion (optical flow) from a monocular video without any explicit supervision from ground truth annotations, particularly hard to source for these two tasks. In this paper, we take an additional step toward holistic scene understanding with monocular cameras by learning depth and motion alongside with semantics, with supervision for the latter provided by a pre-trained network distilling proxy ground truth images. We address the three tasks jointly by a) a novel training protocol based on knowledge distillation and self-supervision and b) a compact network architecture which enables efficient scene understanding on both power hungry GPUs and low-power embedded platforms. We thoroughly assess the performance of our framework and show that it yields state-of-the-art results for monocular depth estimation, optical flow and motion segmentation.

Deep architectures represent the state-of-the-art for perceiving depth from stereo images. Although these methods are highly accurate, it is crucial to effectively detect any outlier through confidence measures since a wrong perception of even small portions of the sensed scene might lead to catastrophic consequences, for instance, in autonomous driving. Purposely, state-of-the-art confidence estimation methods rely on deep-networks as well. In this paper, arguing that these tasks are two sides of the same coin, we propose a novel paradigm for their joint training. Specifically, inspired by the successful deployment of GANs in other fields, we design two deep architectures: a generator for disparity estimation and a discriminator for distinguishing correct assignments from outliers. The two networks are jointly trained in a new peculiar weakly adversarial manner pushing the former to fix the errors detected by the discriminator while keeping the correct prediction unchanged. Experimental results on standard stereo datasets prove that such joint training paradigm is beneficial. Moreover, an additional outcome of our proposal is the ability to detect outliers with better accuracy compared to the state-of-the-art.

Stereo is a prominent technique to infer dense depth maps from images, and deep learning further pushed forward the state-of-the-art, making end-to-end architectures unrivaled when enough data is available for training. However, deep networks suffer from significant drops in accuracy when dealing with new environments. Therefore, in this paper, we introduce Guided Stereo Matching, a novel paradigm leveraging a small amount of sparse, yet reliable depth measurements retrieved from an external source enabling to ameliorate this weakness. The additional sparse cues required by our method can be obtained with any strategy (e.g., a LiDAR) and used to enhance features linked to corresponding disparity hypotheses. Our formulation is general and fully differentiable, thus enabling to exploit the additional sparse inputs in pre-trained deep stereo networks as well as for training a new instance from scratch. Extensive experiments on three standard datasets and two state-of-the-art deep architectures show that even with a small set of sparse input cues, i) the proposed paradigm enables significant improvements to pre-trained networks. Moreover, ii) training from scratch notably increases accuracy and robustness to domain shifts. Finally, iii) it is suited and effective even with traditional stereo algorithms such as SGM.

Deep convolutional neural networks trained end-to-end are the state-of-the-art methods to regress dense disparity maps from stereo pairs. These models, however, suffer from a notable decrease in accuracy when exposed to scenarios significantly different from the training set (e.g., real vs synthetic images, etc.). We argue that it is extremely unlikely to gather enough samples to achieve effective training/tuning in any target domain, thus making this setup impractical for many applications. Instead, we propose to perform unsupervised and continuous online adaptation of a deep stereo network, which allows for preserving its accuracy in any environment. However, this strategy is extremely computationally demanding and thus prevents real-time inference. We address this issue introducing a new lightweight, yet effective, deep stereo architecture, Modularly ADaptive Network(MADNet), and developing a Modular ADaptation (MAD) algorithm, which independently trains sub-portions of the network. By deploying MADNet together with MAD we introduce the first real-time self-adaptive deep stereo system enabling competitive performance on heterogeneous datasets. Our code is publicly available at https://github.com/CVLAB-Unibo/Real-time-self-adaptive-deep-stereo.

Depth estimation from a single image represents a fascinating, yet challenging problem with countless applications. Recent works proved that this task could be learned without direct supervision from ground truth labels leveraging image synthesis on sequences or stereo pairs. Focusing on this second case, in this paper we leverage stereo matching in order to improve monocular depth estimation. To this aim we propose monoResMatch, a novel deep architecture designed to infer depth from a single input image by synthesizing features from a different point of view, horizontally aligned with the input image, performing stereo matching between the two cues. In contrast to previous works sharing this rationale, our network is the first trained end-to-end from scratch. Moreover, we show how obtaining proxy ground truth annotation through traditional stereo algorithms, such as Semi-Global Matching, enables more accurate monocular depth estimation still countering the need for expensive depth labels by keeping a self-supervised approach. Exhaustive experimental results prove how the synergy between i) the proposed monoResMatch architecture and ii) proxy-supervision attains state-of-the-art for self-supervised monocular depth estimation. The code is publicly available at https://github.com/fabiotosi92/monoResMatch-Tensorflow.

This work deals with the implementation of energy-efficient monocular depth estimation using a low-cost CPU for low-power embedded systems. It first describes the PyD-Net depth estimation network, which consists of a lightweight CNN able to approach state-of-the-art accuracy with ultra-low resource usage. Then it proposes an accuracy-driven complexity reduction strategy based on a hardware-friendly fixed-point quantization. Finally, it introduces the low-level optimization enabling effective use of integer neural kernels. The objective is threefold: (i) prove the efficiency of the new quantization flow on a depth estimation network, that is, the capability to retaining the accuracy reached by floating-point arithmetic using 16- and 8-bit integers, (ii) demonstrate the portability of the quantized model into a general-purpose 32-bit RISC architecture of the ARM Cortex family, (iii) quantify the accuracy-energy tradeoff of unsupervised monocular estimation to establish its use in the embedded domain. The experiments have been run on a Raspberry PI board powered by a Broadcom BCM2837 chipset. A parametric analysis conducted over the KITTI date-set shows marginal accuracy loss with 16-bit (8-bit) integers and energy savings up to 6.55× (9.23×) w.r.t. floating-point. Compared to high-end CPU and GPU the proposed solution improves scalability.

Obtaining accurate depth measurements out of a single image represents a fascinating solution to 3D sensing. CNNs led to considerable improvements in this field, and recent trends replaced the need for ground-truth labels with geometry-guided image reconstruction signals enabling unsupervised training. Currently, for this purpose, state-of-the-art techniques rely on images acquired with a binocular stereo rig to predict inverse depth (i.e., disparity) according to the aforementioned supervision principle. However, these methods suffer from well-known problems near occlusions, left image border, etc inherited from the stereo setup. Therefore, in this paper, we tackle these issues by moving to a trinocular domain for training. Assuming the central image as the reference, we train a CNN to infer disparity representations pairing such image with frames on its left and right side. This strategy allows obtaining depth maps not affected by typical stereo artifacts. Moreover, being trinocular datasets seldom available, we introduce a novel interleaved training procedure enabling to enforce the trinocular assumption outlined from current binocular datasets. Exhaustive experimental results on the KITTI dataset confirm that our proposal outperforms state-of-the-art methods for unsupervised monocular depth estimation trained on binocular stereo pairs as well as any known methods relying on other cues.

Unsupervised depth estimation from a single image is a very attractive technique with several implications in robotic, autonomous navigation, augmented reality and so on. This topic represents a very challenging task and the advent of deep learning enabled to tackle this problem with excellent results. However, these architectures are extremely deep and complex. Thus, real-time performance can be achieved only by leveraging power-hungry GPUs that do not allow to infer depth maps in application fields characterized by low-power constraints. To tackle this issue, in this paper we propose a novel architecture capable to quickly infer an accurate depth map on a CPU, even of an embedded system, using a pyramid of features extracted from a single input image. Similarly to state-of-the-art, we train our network in an unsupervised manner casting depth estimation as an image reconstruction problem. Extensive experimental results on the KITTI dataset show that compared to the top performing approach our network has similar accuracy but a much lower complexity (about 6% of parameters) enabling to infer a depth map for a KITTI image in about 1.7 s on the Raspberry Pi 3 and at more than 8 Hz on a standard CPU. Moreover, by trading accuracy for efficiency, our network allows to infer maps at about 2 Hz and 40 Hz respectively, still being more accurate than most state-of-the-art slower methods. To the best of our knowledge, it is the first method enabling such performance on CPUs paving the way for effective deployment of unsupervised monocular depth estimation even on embedded systems.

Confidence measures aim at detecting unreliable depth measurements and play an important role for many purposes and in particular, as recently shown, to improve stereo accuracy. This topic has been thoroughly investigated by Hu and Mordohai in 2010 (and 2012) considering 17 confidence measures and two local algorithms on the two datasets available at that time. However, since then major breakthroughs happened in this field: the availability of much larger and challenging datasets, novel and more effective stereo algorithms including ones based on deep learning and confidence measures leveraging on machine learning techniques. Therefore, this paper aims at providing an exhaustive and updated review and quantitative evaluation of 52 (actually, 76 considering variants) state-of-the-art confidence measures - focusing on recent ones mostly based on random-forests and deep learning - with three algorithms on the challenging datasets available today. Moreover we deal with problems inherently induced by learning-based confidence measures. How are these methods able to generalize to new data? How a specific training improves their effectiveness? How more effective confidence measures can actually improve the overall stereo accurac?