PyTorchKR Papers

최신 논문

총 2,547편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1707.09603

Occlusion Handling using Semantic Segmentation and Visibility-Based Rendering for Mixed Reality

Real-time occlusion handling is a major problem in outdoor mixed reality system because it requires great computational cost mainly due to the complexity of the scene. Using only segmentation, it is difficult to accurately render a virtual object occluded by complex objects such as trees, bushes etc. In this paper, we propose a novel occlusion handling method for real-time, outdoor, and omni-directional mixed reality system using only the information from a monocular image sequence. We first present a semantic segmentation scheme for predicting the amount of visibility for different type of objects in the scene. We also simultaneously calculate a foreground probability map using depth estimation derived from optical flow. Finally, we combine the segmentation result and the probability map to render the computer generated object and the real scene using a visibility-based rendering method. Our results show great improvement in handling occlusions compared to existing blending based methods.

arXiv:1701.02141

Light Field Super-Resolution Via Graph-Based Regularization

Light field cameras capture the 3D information in a scene with a single exposure. This special feature makes light field cameras very appealing for a variety of applications: from post-capture refocus, to depth estimation and image-based rendering. However, light field cameras suffer by design from strong limitations in their spatial resolution, which should therefore be augmented by computational methods. On the one hand, off-the-shelf single-frame and multi-frame super-resolution algorithms are not ideal for light field data, as they do not consider its particular structure. On the other hand, the few super-resolution algorithms explicitly tailored for light field data exhibit significant limitations, such as the need to estimate an explicit disparity map at each view. In this work we propose a new light field super-resolution algorithm meant to address these limitations. We adopt a multi-frame alike super-resolution approach, where the complementary information in the different light field views is used to augment the spatial resolution of the whole light field. We show that coupling the multi-frame approach with a graph regularizer, that enforces the light field structure via nonlocal self similarities, permits to avoid the costly and challenging disparity estimation step for all the views. Extensive experiments show that the new algorithm compares favorably to the other state-of-the-art methods for light field super-resolution, both in terms of PSNR and visual quality.

arXiv:1611.05008

Hybrid Light Field Imaging for Improved Spatial Resolution and Depth Range

Light field imaging involves capturing both angular and spatial distribution of light; it enables new capabilities, such as post-capture digital refocusing, camera aperture adjustment, perspective shift, and depth estimation. Micro-lens array (MLA) based light field cameras provide a cost-effective approach to light field imaging. There are two main limitations of MLA-based light field cameras: low spatial resolution and narrow baseline. While low spatial resolution limits the general purpose use and applicability of light field cameras, narrow baseline limits the depth estimation range and accuracy. In this paper, we present a hybrid stereo imaging system that includes a light field camera and a regular camera. The hybrid system addresses both spatial resolution and narrow baseline issues of the MLA-based light field cameras while preserving light field imaging capabilities.

arXiv:1708.02287

Monocular Depth Estimation with Hierarchical Fusion of Dilated CNNs and Soft-Weighted-Sum Inference

Monocular depth estimation is a challenging task in complex compositions depicting multiple objects of diverse scales. Albeit the recent great progress thanks to the deep convolutional neural networks (CNNs), the state-of-the-art monocular depth estimation methods still fall short to handle such real-world challenging scenarios. In this paper, we propose a deep end-to-end learning framework to tackle these challenges, which learns the direct mapping from a color image to the corresponding depth map. First, we represent monocular depth estimation as a multi-category dense labeling task by contrast to the regression based formulation. In this way, we could build upon the recent progress in dense labeling such as semantic segmentation. Second, we fuse different side-outputs from our front-end dilated convolutional neural network in a hierarchical way to exploit the multi-scale depth cues for depth estimation, which is critical to achieve scale-aware depth estimation. Third, we propose to utilize soft-weighted-sum inference instead of the hard-max inference, transforming the discretized depth score to continuous depth value. Thus, we reduce the influence of quantization error and improve the robustness of our method. Extensive experiments on the NYU Depth V2 and KITTI datasets show the superiority of our method compared with current state-of-the-art methods. Furthermore, experiments on the NYU V2 dataset reveal that our model is able to learn the probability distribution of depth.

arXiv:1704.07813공식 코드 · Papers with Code

Unsupervised Learning of Depth and Ego-Motion from Video

We present an unsupervised learning framework for the task of monocular depth and camera motion estimation from unstructured video sequences. We achieve this by simultaneously training depth and camera pose estimation networks using the task of view synthesis as the supervisory signal. The networks are thus coupled via the view synthesis objective during training, but can be applied independently at test time. Empirical evaluation on the KITTI dataset demonstrates the effectiveness of our approach: 1) monocular depth performing comparably with supervised methods that use either ground-truth pose or depth for training, and 2) pose estimation performing favorably with established SLAM systems under comparable input settings.

arXiv:1708.01964

Accurate Light Field Depth Estimation with Superpixel Regularization over Partially Occluded Regions

Depth estimation is a fundamental problem for light field photography applications. Numerous methods have been proposed in recent years, which either focus on crafting cost terms for more robust matching, or on analyzing the geometry of scene structures embedded in the epipolar-plane images. Significant improvements have been made in terms of overall depth estimation error; however, current state-of-the-art methods still show limitations in handling intricate occluding structures and complex scenes with multiple occlusions. To address these challenging issues, we propose a very effective depth estimation framework which focuses on regularizing the initial label confidence map and edge strength weights. Specifically, we first detect partially occluded boundary regions (POBR) via superpixel based regularization. Series of shrinkage/reinforcement operations are then applied on the label confidence map and edge strength weights over the POBR. We show that after weight manipulations, even a low-complexity weighted least squares model can produce much better depth estimation than state-of-the-art methods in terms of average disparity error rate, occlusion boundary precision-recall rate, and the preservation of intricate visual features.

arXiv:1708.02837

SPLODE: Semi-Probabilistic Point and Line Odometry with Depth Estimation from RGB-D Camera Motion

Active depth cameras suffer from several limitations, which cause incomplete and noisy depth maps, and may consequently affect the performance of RGB-D Odometry. To address this issue, this paper presents a visual odometry method based on point and line features that leverages both measurements from a depth sensor and depth estimates from camera motion. Depth estimates are generated continuously by a probabilistic depth estimation framework for both types of features to compensate for the lack of depth measurements and inaccurate feature depth associations. The framework models explicitly the uncertainty of triangulating depth from both point and line observations to validate and obtain precise estimates. Furthermore, depth measurements are exploited by propagating them through a depth map registration module and using a frame-to-frame motion estimation method that considers 3D-to-2D and 2D-to-3D reprojection errors, independently. Results on RGB-D sequences captured on large indoor and outdoor scenes, where depth sensor limitations are critical, show that the combination of depth measurements and estimates through our approach is able to overcome the absence and inaccuracy of depth measurements.

arXiv:1708.02550

Fast Scene Understanding for Autonomous Driving

Most approaches for instance-aware semantic labeling traditionally focus on accuracy. Other aspects like runtime and memory footprint are arguably as important for real-time applications such as autonomous driving. Motivated by this observation and inspired by recent works that tackle multiple tasks with a single integrated architecture, in this paper we present a real-time efficient implementation based on ENet that solves three autonomous driving related tasks at once: semantic scene segmentation, instance segmentation and monocular depth estimation. Our approach builds upon a branched ENet architecture with a shared encoder but different decoder branches for each of the three tasks. The presented method can run at 21 fps at a resolution of 1024x512 on the Cityscapes dataset without sacrificing accuracy compared to running each task separately.

arXiv:1708.03292

Learning to Synthesize a 4D RGBD Light Field from a Single Image

We present a machine learning algorithm that takes as input a 2D RGB image and synthesizes a 4D RGBD light field (color and depth of the scene in each ray direction). For training, we introduce the largest public light field dataset, consisting of over 3300 plenoptic camera light fields of scenes containing flowers and plants. Our synthesis pipeline consists of a convolutional neural network (CNN) that estimates scene geometry, a stage that renders a Lambertian light field using that geometry, and a second CNN that predicts occluded rays and non-Lambertian effects. Our algorithm builds on recent view synthesis methods, but is unique in predicting RGBD for each light field ray and improving unsupervised single image depth estimation by enforcing consistency of ray depths that should intersect the same scene point. Please see our supplementary video at https://youtu.be/yLCvWoQLnms

arXiv:1605.02305

Estimating Depth from Monocular Images as Classification Using Deep Fully Convolutional Residual Networks

Depth estimation from single monocular images is a key component of scene understanding and has benefited largely from deep convolutional neural networks (CNN) recently. In this article, we take advantage of the recent deep residual networks and propose a simple yet effective approach to this problem. We formulate depth estimation as a pixel-wise classification task. Specifically, we first discretize the continuous depth values into multiple bins and label the bins according to their depth range. Then we train fully convolutional deep residual networks to predict the depth label of each pixel. Performing discrete depth label classification instead of continuous depth value regression allows us to predict a confidence in the form of probability distribution. We further apply fully-connected conditional random fields (CRF) as a post processing step to enforce local smoothness interactions, which improves the results. We evaluate our approach on both indoor and outdoor datasets and achieve state-of-the-art performance.

arXiv:1707.05847

The Devil is in the Decoder: Classification, Regression and GANs

Many machine vision applications, such as semantic segmentation and depth prediction, require predictions for every pixel of the input image. Models for such problems usually consist of encoders which decrease spatial resolution while learning a high-dimensional representation, followed by decoders who recover the original input resolution and result in low-dimensional predictions. While encoders have been studied rigorously, relatively few studies address the decoder side. This paper presents an extensive comparison of a variety of decoders for a variety of pixel-wise tasks ranging from classification, regression to synthesis. Our contributions are: (1) Decoders matter: we observe significant variance in results between different types of decoders on various problems. (2) We introduce new residual-like connections for decoders. (3) We introduce a novel decoder: bilinear additive upsampling. (4) We explore prediction artifacts.

arXiv:1708.06779

Reflection Separation and Deblurring of Plenoptic Images

In this paper, we address the problem of reflection removal and deblurring from a single image captured by a plenoptic camera. We develop a two-stage approach to recover the scene depth and high resolution textures of the reflected and transmitted layers. For depth estimation in the presence of reflections, we train a classifier through convolutional neural networks. For recovering high resolution textures, we assume that the scene is composed of planar regions and perform the reconstruction of each layer by using an explicit form of the plenoptic camera point spread function. The proposed framework also recovers the sharp scene texture with different motion blurs applied to each layer. We demonstrate our method on challenging real and synthetic images.

arXiv:1708.07860

Multi-task Self-Supervised Visual Learning

We investigate methods for combining multiple self-supervised tasks--i.e., supervised tasks where data can be collected without manual labeling--in order to train a single visual representation. First, we provide an apples-to-apples comparison of four different self-supervised tasks using the very deep ResNet-101 architecture. We then combine tasks to jointly train a network. We also explore lasso regularization to encourage the network to factorize the information in its representation, and methods for "harmonizing" network inputs in order to learn a more unified representation. We evaluate all methods on ImageNet classification, PASCAL VOC detection, and NYU depth prediction. Our results show that deeper networks work better, and that combining tasks--even via a naive multi-head architecture--always improves performance. Our best joint network nearly matches the PASCAL performance of a model pre-trained on ImageNet classification, and matches the ImageNet network on NYU depth prediction.

arXiv:1709.00072

Exact Blur Measure Outperforms Conventional Learned Features for Depth Finding

Image analysis methods that are based on exact blur values are faced with the computational complexities due to blur measurement error. This atmosphere encourages scholars to look for handcrafted and learned features for finding depth from a single image. This paper introduces a novel exact realization for blur measures on digital images and implements it on a new measure of defocus Gaussian blur at edge points in Depth From Defocus (DFD) methods with the potential to change this atmosphere. The experiments on real images indicate superiority of the proposed measure in error performance over conventional learned features in the state-of the-art single image based depth estimation methods.

arXiv:1708.06500

Sparsity Invariant CNNs

In this paper, we consider convolutional neural networks operating on sparse inputs with an application to depth upsampling from sparse laser scan data. First, we show that traditional convolutional networks perform poorly when applied to sparse data even when the location of missing data is provided to the network. To overcome this problem, we propose a simple yet effective sparse convolution layer which explicitly considers the location of missing data during the convolution operation. We demonstrate the benefits of the proposed network architecture in synthetic and real experiments with respect to various baseline approaches. Compared to dense baselines, the proposed sparse convolution network generalizes well to novel datasets and is invariant to the level of sparsity in the data. For our evaluation, we derive a novel dataset from the KITTI benchmark, comprising 93k depth annotated RGB images. Our dataset allows for training and evaluating depth upsampling and depth prediction techniques in challenging real-world settings and will be made available upon publication.

arXiv:1709.07077공식 코드 · Papers with Code

Estimated Depth Map Helps Image Classification

We consider image classification with estimated depth. This problem falls into the domain of transfer learning, since we are using a model trained on a set of depth images to generate depth maps (additional features) for use in another classification problem using another disjoint set of images. It's challenging as no direct depth information is provided. Though depth estimation has been well studied, none have attempted to aid image classification with estimated depth. Therefore, we present a way of transferring domain knowledge on depth estimation to a separate image classification task over a disjoint set of train, and test data. We build a RGBD dataset based on RGB dataset and do image classification on it. Then evaluation the performance of neural networks on the RGBD dataset compared to the RGB dataset. From our experiments, the benefit is significant with shallow and deep networks. It improves ResNet-20 by 0.55% and ResNet-56 by 0.53%. Our code and dataset are available publicly.

arXiv:1710.00513

Depth estimation using structured light flow -- analysis of projected pattern flow on an object's surface --

Shape reconstruction techniques using structured light have been widely researched and developed due to their robustness, high precision, and density. Because the techniques are based on decoding a pattern to find correspondences, it implicitly requires that the projected patterns be clearly captured by an image sensor, i.e., to avoid defocus and motion blur of the projected pattern. Although intensive researches have been conducted for solving defocus blur, few researches for motion blur and only solution is to capture with extremely fast shutter speed. In this paper, unlike the previous approaches, we actively utilize motion blur, which we refer to as a light flow, to estimate depth. Analysis reveals that minimum two light flows, which are retrieved from two projected patterns on the object, are required for depth estimation. To retrieve two light flows at the same time, two sets of parallel line patterns are illuminated from two video projectors and the size of motion blur of each line is precisely measured. By analyzing the light flows, i.e. lengths of the blurs, scene depth information is estimated. In the experiments, 3D shapes of fast moving objects, which are inevitably captured with motion blur, are successfully reconstructed by our technique.

arXiv:1703.04977

What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?

There are two major types of uncertainty one can model. Aleatoric uncertainty captures noise inherent in the observations. On the other hand, epistemic uncertainty accounts for uncertainty in the model -- uncertainty which can be explained away given enough data. Traditionally it has been difficult to model epistemic uncertainty in computer vision, but with new Bayesian deep learning tools this is now possible. We study the benefits of modeling epistemic vs. aleatoric uncertainty in Bayesian deep learning models for vision tasks. For this we present a Bayesian deep learning framework combining input-dependent aleatoric uncertainty together with epistemic uncertainty. We study models under the framework with per-pixel semantic segmentation and depth regression tasks. Further, our explicit uncertainty formulation leads to new loss functions for these tasks, which can be interpreted as learned attenuation. This makes the loss more robust to noisy data, also giving new state-of-the-art results on segmentation and depth regression benchmarks.

arXiv:1710.05221

An Adaptive Framework for Missing Depth Inference Using Joint Bilateral Filter

Depth imaging has largely focused on sensor and intrinsics properties. However, the accuracy of acquire pixel is largely dependent on the capture. We propose a new depth estimation and approximation algorithm which takes an arbitrary 3D point cloud as input, with potentially complex geometric structures, and generates automatically a bounding box which is used to clamp the 3D distribution into a valid range. We then infer the desired compact geometric network from complex 3D geometries by using a series of adaptive joint bilateral filters. Our approach leverages these input depth in the construction of a compact descriptive adaptive filter framework. The built system that allows a user to control the result of capture depth map to fit the target geometry. In addition, it is desirable to visualize structurally problematic areas of the depth data in a dynamic environment. To provide this feature, we investigate a fast update algorithm for the fragility of each pixel's corresponding 3D point using machine learning. We present a new for of feature vector analysis and demonstrate the effectiveness in the dataset. In our experiment, we demonstrate the practicality and benefits of our proposed method by computing accurate solutions captured depth map from different types of sensors and shows better results than existing methods.

arXiv:1703.01398

Sparse Depth Sensing for Resource-Constrained Robots

We consider the case in which a robot has to navigate in an unknown environment but does not have enough on-board power or payload to carry a traditional depth sensor (e.g., a 3D lidar) and thus can only acquire a few (point-wise) depth measurements. We address the following question: is it possible to reconstruct the geometry of an unknown environment using sparse and incomplete depth measurements? Reconstruction from incomplete data is not possible in general, but when the robot operates in man-made environments, the depth exhibits some regularity (e.g., many planar surfaces with only a few edges); we leverage this regularity to infer depth from a small number of measurements. Our first contribution is a formulation of the depth reconstruction problem that bridges robot perception with the compressive sensing literature in signal processing. The second contribution includes a set of formal results that ascertain the exactness and stability of the depth reconstruction in 2D and 3D problems, and completely characterize the geometry of the profiles that we can reconstruct. Our third contribution is a set of practical algorithms for depth reconstruction: our formulation directly translates into algorithms for depth estimation based on convex programming. In real-world problems, these convex programs are very large and general-purpose solvers are relatively slow. For this reason, we discuss ad-hoc solvers that enable fast depth reconstruction in real problems. The last contribution is an extensive experimental evaluation in 2D and 3D problems, including Monte Carlo runs on simulated instances and testing on multiple real datasets. Empirical results confirm that the proposed approach ensures accurate depth reconstruction, outperforms interpolation-based strategies, and performs well even when the assumption of structured environment is violated.