RESUMO
In this paper, we propose a unified and flexible framework for general image fusion tasks, including multi-exposure image fusion, multi-focus image fusion, infrared/visible image fusion, and multi-modality medical image fusion. Unlike other deep learning-based image fusion methods applied to a fixed number of input sources (normally two inputs), the proposed framework can simultaneously handle an arbitrary number of inputs. Specifically, we use the symmetrical function (e.g., Max-pooling) to extract the most significant features from all the input images, which are then fused with the respective features from each input source. This symmetry function enables permutation-invariance of the network, which means the network can successfully extract and fuse the saliency features of each image without needing to remember the input order of the inputs. The property of permutation-invariance also brings convenience for the network during inference with unfixed inputs. To handle multiple image fusion tasks with one unified framework, we adopt continual learning based on Elastic Weight Consolidation (EWC) for different fusion tasks. Subjective and objective experiments on several public datasets demonstrate that the proposed method outperforms state-of-the-art methods on multiple image fusion tasks.
Assuntos
Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Processamento de Imagem Assistida por Computador/métodos , RegistrosRESUMO
Multi-exposure image fusion (MEF) methods for high dynamic range (HDR) imaging suffer from ghosting artifacts when dealing with moving objects in dynamic scenes. The state-of-the-art methods use optical flow to align low dynamic range (LDR) images before merging, introducing distortion into the aligned LDR images from inaccurate motion estimation due to large motion and occlusion. In place of pre-alignment, attention-based methods calculate the correlation between the reference LDR image and non-reference LDR images, thus excluding misaligned regions in LDR images. Nevertheless, they also exclude the saturated details at the same time. Taking advantage of both the alignment and attention-based methods, we propose an efficient Deep HDR Deghosting Fusion Network (DDFNet) guided by optical flow and image correlation attentions. Specifically, the DDFNet estimates the optical flow of the LDR images by a motion estimation module and encodes that optical flow as a flow feature. Additionally, it extracts correlation features between the reference LDR and other non-reference LDR images. The optical flow and correlation features are employed to adaptably combine information from LDR inputs in an attention-based fusion module. Following the merging of features, a decoder composed of Dense Networks reconstructs the HDR image without ghosting. Experimental results indicate that the proposed DDFNet achieves state-of-the-art image fusion performance on different public datasets.
Assuntos
Artefatos , Movimento (Física)RESUMO
Pedestrian detection is an important research domain due to its relevance for autonomous and assisted driving, as well as its applications in security and industrial automation. Often, more than one type of sensor is used to cover a broader range of operating conditions than a single-sensor system would allow. However, it remains difficult to make pedestrian detection systems perform well in highly dynamic environments, often requiring extensive retraining of the algorithms for specific conditions to reach satisfactory accuracy, which, in turn, requires large, annotated datasets captured in these conditions. In this paper, we propose a probabilistic decision-level sensor fusion method based on naive Bayes to improve the efficiency of the system by combining the output of available pedestrian detectors for colour and thermal images without retraining. The results in this paper, obtained through long-term experiments, demonstrate the efficacy of our technique, its ability to work with non-registered images, and its adaptability to cope with situations when one of the sensors fails. The results also show that our proposed technique improves the overall accuracy of the system and could be very useful in several applications.
Assuntos
Condução de Veículo , Pedestres , Humanos , Teorema de Bayes , Cor , AlgoritmosRESUMO
Depth sensing has improved rapidly in recent years, which allows for structural information to be utilized in various applications, such as virtual reality, scene and object recognition, view synthesis, and 3D reconstruction. Due to the limitations of the current generation of depth sensors, the resolution of depth maps is often still much lower than the resolution of color images. This hinders applications, such as view synthesis or 3D reconstruction, from providing high-quality results. Therefore, super-resolution, which allows for the upscaling of depth maps while still retaining sharpness, has recently drawn much attention in the deep learning community. However, state-of-the-art deep learning methods are typically designed and trained to handle a fixed set of integer-scale factors. Moreover, the raw depth map collected by the depth sensor usually has many depth data missing or misestimated values along the edges and corners of observed objects. In this work, we propose a novel deep learning network for both depth completion and depth super-resolution with arbitrary scale factors. The experimental results on the Middlebury stereo, NYUv2, and Matterport3D datasets demonstrate that the proposed method can outperform state-of-the-art methods.
Assuntos
Realidade VirtualRESUMO
Drafting involves cycling so close behind another person that wind resistance is significantly reduced, which is illegal during most long distance and several short distance triathlon and duathlon events. In this paper, a proof of concept for a drafting detection system based on computer vision is proposed. After detecting and tracking a bicycle through the various scenes, the distance to this object is estimated through computational geometry. The probability of drafting is then determined through statistical analysis of subsequent measurements over an extended period of time. These algorithms are tested using a static recording and a recording that simulates a race situation with ground truth distances obtained from a Light Detection And Ranging (LiDAR) system. The most accurate developed distance estimation method yields an average error of 0 . 46 m in our test scenario. When sampling the distances at periods of 1 or 2 s, simulations demonstrate that a drafting violation is detected quickly for cyclists riding at 2 m or more below the limit, while generally avoiding false positives during the race-like test set-up and five hour race simulations.
Assuntos
Ciclismo , Fotografação/instrumentação , Vento , Algoritmos , Simulação por Computador , Humanos , ProbabilidadeRESUMO
This paper presents a vulnerable road user (VRU) tracking algorithm capable of handling noisy and missing detections from heterogeneous sensors. We propose a cooperative fusion algorithm for matching and reinforcing of radar and camera detections using their proximity and positional uncertainty. The belief in the existence and position of objects is then maximized by temporal integration of fused detections by a multi-object tracker. By switching between observation models, the tracker adapts to the detection noise characteristics making it robust to individual sensor failures. The main novelty of this paper is an improved imputation sampling function for updating the state when detections are missing. The proposed function uses a likelihood without association that is conditioned on the sensor information instead of the sensor model. The benefits of the proposed solution are two-fold: firstly, particle updates become computationally tractable and secondly, the problem of imputing samples from a state which is predicted without an associated detection is bypassed. Experimental evaluation shows a significant improvement in both detection and tracking performance over multiple control algorithms. In low light situations, the cooperative fusion outperforms intermediate fusion by as much as 30%, while increases in tracking performance are most significant in complex traffic scenes.
RESUMO
In this paper, we present a novel 2Dâ»3D pedestrian tracker designed for applications in autonomous vehicles. The system operates on a tracking by detection principle and can track multiple pedestrians in complex urban traffic situations. By using a behavioral motion model and a non-parametric distribution as state model, we are able to accurately track unpredictable pedestrian motion in the presence of heavy occlusion. Tracking is performed independently, on the image and ground plane, in global, motion compensated coordinates. We employ Camera and LiDAR data fusion to solve the association problem where the optimal solution is found by matching 2D and 3D detections to tracks using a joint log-likelihood observation model. Each 2Dâ»3D particle filter then updates their state from associated observations and a behavioral motion model. Each particle moves independently following the pedestrian motion parameters which we learned offline from an annotated training dataset. Temporal stability of the state variables is achieved by modeling each track as a Markov Decision Process with probabilistic state transition properties. A novel track management system then handles high level actions such as track creation, deletion and interaction. Using a probabilistic track score the track manager can cull false and ambiguous detections while updating tracks with detections from actual pedestrians. Our system is implemented on a GPU and exploits the massively parallelizable nature of particle filters. Due to the Markovian nature of our track representation, the system achieves real-time performance operating with a minimal memory footprint. Exhaustive and independent evaluation of our tracker was performed by the KITTI benchmark server, where it was tested against a wide variety of unknown pedestrian tracking situations. On this realistic benchmark, we outperform all published pedestrian trackers in a multitude of tracking metrics.
RESUMO
In this paper, we describe a robust method for compensating the panning and tilting motion of a camera, applied to foreground-background segmentation. First, the necessary internal camera parameters are determined through feature-point extraction and tracking. From these parameters, two motion models for points in the image plane are established. The first model assumes a fixed tilt angle, whereas the second model allows simultaneous pan and tilt. At runtime, these models are used to compensate for the motion of the camera in the background model. We will show that these methods provide a robust compensation mechanism and improve the foreground masks of an otherwise state-of-the-art unsupervised foreground-background segmentation method. The resulting algorithm is always able to obtain F 1 scores above 80 % on every daytime video in our test set when a minimal number of only eight feature matches are used to determine the background compensation, whereas the standard approaches need significantly more feature matches to produce similar results.
RESUMO
Better features have been driving the progress of pedestrian detection over the past years. However, as features become richer and higher dimensional, noise and redundancy in the feature sets become bigger problems. These problems slow down learning and can even reduce the performance of the learned model. Current solutions typically exploit dimension reduction techniques. In this paper, we propose a simple but effective feature selection framework for pedestrian detection. Moreover, we introduce occluded pedestrian samples into the training process and combine it with a new feature selection criterion, which enables improved performances for occlusion handling problems. Experimental results on the Caltech Pedestrian dataset demonstrate the efficiency of our method over the state-of-art methods, especially for the occluded pedestrians.
RESUMO
In this paper, we present a forward collision warning application for smartphones that uses license plate recognition and vehicular communication to generate warnings for notifying the drivers of the vehicle behind and the one ahead, of a probable collision when the vehicle behind does not maintain an established safe distance between itself and the vehicle ahead. The application was tested in both static and mobile scenarios, from which we confirmed the working of our application, even though its performance is affected by the hardware limitations of the smartphones.
RESUMO
In this paper, we propose a novel extrinsic calibration method for camera networks by analyzing tracks of pedestrians. First of all, we extract the center lines of walking persons by detecting their heads and feet in the camera images. We propose an easy and accurate method to estimate the 3D positions of the head and feet w.r.t. a local camera coordinate system from these center lines. We also propose a RANSAC-based orthogonal Procrustes approach to compute relative extrinsic parameters connecting the coordinate systems of cameras in a pairwise fashion. Finally, we refine the extrinsic calibration matrices using a method that minimizes the reprojection error. While existing state-of-the-art calibration methods explore epipolar geometry and use image positions directly, the proposed method first computes 3D positions per camera and then fuses the data. This results in simpler computations and a more flexible and accurate calibration method. Another advantage of our method is that it can also handle the case of persons walking along straight lines, which cannot be handled by most of the existing state-of-the-art calibration methods since all head and feet positions are co-planar. This situation often happens in real life.
RESUMO
In this paper, we propose a novel extrinsic calibration method for camera networks using a sphere as the calibration object. First of all, we propose an easy and accurate method to estimate the 3D positions of the sphere center w.r.t. the local camera coordinate system. Then, we propose to use orthogonal procrustes analysis to pairwise estimate the initial camera relative extrinsic parameters based on the aforementioned estimation of 3D positions. Finally, an optimization routine is applied to jointly refine the extrinsic parameters for all cameras. Compared to existing sphere-based 3D position estimators which need to trace and analyse the outline of the sphere projection in the image, the proposed method requires only very simple image processing: estimating the area and the center of mass of the sphere projection. Our results demonstrate that we can get a more accurate estimate of the extrinsic parameters compared to other sphere-based methods. While existing state-of-the-art calibration methods use point like features and epipolar geometry, the proposed method uses the sphere-based 3D position estimate. This results in simpler computations and a more flexible and accurate calibration method. Experimental results show that the proposed approach is accurate, robust, flexible and easy to use.
RESUMO
This paper proposes an automated system for monitoring mobility patterns using a network of very low resolution visual sensors (30 × 30 pixels). The use of very low resolution sensors reduces privacy concern, cost, computation requirement and power consumption. The core of our proposed system is a robust people tracker that uses low resolution videos provided by the visual sensor network. The distributed processing architecture of our tracking system allows all image processing tasks to be done on the digital signal controller in each visual sensor. In this paper, we experimentally show that reliable tracking of people is possible using very low resolution imagery. We also compare the performance of our tracker against a state-of-the-art tracking method and show that our method outperforms. Moreover, the mobility statistics of tracks such as total distance traveled and average speed derived from trajectories are compared with those derived from ground truth given by Ultra-Wide Band sensors. The results of this comparison show that the trajectories from our system are accurate enough to obtain useful mobility statistics.
Assuntos
Actigrafia/instrumentação , Redes de Comunicação de Computadores/instrumentação , Interpretação de Imagem Assistida por Computador/instrumentação , Atividade Motora/fisiologia , Fotografação/instrumentação , Caminhada/fisiologia , Imagem Corporal Total/instrumentação , Actigrafia/métodos , Desenho de Equipamento , Análise de Falha de Equipamento , Humanos , Interpretação de Imagem Assistida por Computador/métodos , Reconhecimento Automatizado de Padrão/métodos , Fotografação/métodos , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Processamento de Sinais Assistido por Computador/instrumentação , Imagem Corporal Total/métodosRESUMO
This paper presents the design of a multichannel imaging system where the different optical channels have a different angular resolution and field-of-view. Such an imaging system is able to resolve fine details in a small region of interest through the channel that has the highest angular resolution (0.0096°) while controlling the surrounding region through the channel that has the widest field-of-view (2×40°). An interesting feature of such a multichannel, multiresolution imaging system is that various image processing algorithms can be applied at different segments of the image sensor. We have designed a three channel imaging system where each optical channel consists of four aspheric lens surfaces. These three imaging channels share a single image sensor with a resolution of 1440×960 and a 10 µm pixel size. All imaging channels have diffraction-limited performance ensuring good overall image quality.
RESUMO
Although it is generally accepted that visual information guides steering, it is still unclear whether a curvature matching strategy or a 'look where you are going' strategy is used while steering through a curved road. The current experiment investigated to what extent the existing models for curve driving also apply to cycling around a curve, and tested the influence of cycling speed on steering and gaze behavior. Twenty-five participants were asked to cycle through a semicircular lane three consecutive times at three different speeds while staying in the center of the lane. The observed steering behavior suggests that an anticipatory steering strategy was used at curve entrance and a compensatory strategy was used to steer through the actual bend of the curve. A shift of gaze from the center to the inside edge of the lane indicates that at low cycling speed, the 'look where you are going' strategy was preferred, while at higher cycling speeds participants seemed to prefer the curvature matching strategy. Authors suggest that visual information from both steering strategies contributes to the steering system and can be used in a flexible way. Based on a familiarization effect, it can be assumed that steering is not only guided by vision but that a short-term learning component should also be taken into account.