A Survey of Image Gradient Inversion Against Federated Learning

In order to preserve data privacy while fully utilizing data from different owners, federated learning is believed to be a promising approach in recent years. However, aiming at federated learning in the image domain, gradient inversion techniques can reconstruct the input images on pixel-level only by leaked gradients, without accessing the raw data, which makes federated learning vulnerable to the attacks. In this paper, we review the latest advances of image gradient inversion techniques and evaluate the impact of them to federated learning from the attack perspective. We use eight models and four datasets to evaluate the current gradient inversion techniques, comparing the attack performance as well as the time consumption. Furthermore, we shed light on some important and interesting directions of gradient inversion against federated learning.<br>

A Survey of Image Gradient Inversion Against Federated Learning

In order to preserve data privacy while fully utilizing data from different owners, federated learning is believed to be a promising approach in recent years. However, aiming at federated learning in the image domain, gradient inversion techniques can reconstruct the input images on pixel-level only by leaked gradients, without accessing the raw data, which makes federated learning vulnerable to the attacks. In this paper, we review the latest advances of image gradient inversion techniques and evaluate the impact of them to federated learning from the attack perspective. We use eight models and four datasets to evaluate the current gradient inversion techniques, comparing the attack performance as well as the time consumption. Furthermore, we shed light on some important and interesting directions of gradient inversion against federated learning.<br>

A content-adaptive unstructured grid based regularized CT reconstruction method with a SART-type preconditioned fixed-point proximity algorithm

The goal of this study is to develop a new computed tomography (CT) image reconstruction method, aiming at improving the quality of the reconstructed images of existing methods while reducing computational costs. Existing CT reconstruction is modeled by pixel-based piecewise constant approximations of the integral equation that describes the CT projection data acquisition process. Using these approximations imposes a bottleneck model error and results in a discrete system of a large size. We propose to develop a content-adaptive unstructured grid (CAUG) based regularized CT reconstruction method to address these issues. Specifically, we design a CAUG of the image domain to sparsely represent the underlying image, and introduce a CAUG-based piecewise linear approximation of the integral equation by employing a collocation method. We further apply a regularization defined on the CAUG for the resulting illposed linear system, which may lead to a sparse linear representation for the underlying solution. The regularized CT reconstruction is formulated as a convex optimization problem, whose objective function consists of a weighted least square norm based fidelity term, a regularization term and a constraint term. Here, the corresponding weighted matrix is derived from the simultaneous algebraic reconstruction technique (SART). We then develop a SART-type preconditioned fixed-point proximity algorithm to solve the optimization problem. Convergence analysis is provided for the resulting iterative algorithm. Numerical experiments demonstrate the outperformance of the proposed method over several existing methods in terms of both suppressing noise and reducing computational costs. These methods include the SART without regularization and with quadratic regularization on the CAUG, the traditional total variation (TV) regularized reconstruction method and the TV superiorized conjugate gradient method on the pixel grid.

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

The demand for wheelchairs has increased recently as the population of the elderly and patients with disorders increases. However, society still pays less attention to infrastructure that can threaten the wheelchair user, such as sidewalks with cracks/potholes. Although various studies have been proposed to recognize such challenges, they mainly depend on RGB images or IMU sensors, which are sensitive to outdoor conditions such as low illumination, bad weather, and unavoidable vibrations, resulting in unsatisfactory and unstable performance. In this paper, we introduce a novel system based on various convolutional neural networks (CNNs) to automatically classify the condition of sidewalks using images captured with depth and infrared modalities. Moreover, we compare the performance of training CNNs from scratch and the transfer learning approach, where the weights learned from the natural image domain (e.g., ImageNet) are fine-tuned to the depth and infrared image domain. In particular, we propose applying the ResNet-152 model pre-trained with self-supervised learning during transfer learning to leverage better image representations. Performance evaluation on the classification of the sidewalk condition was conducted with 100% and 10% of training data. The experimental results validate the effectiveness and feasibility of the proposed approach and bring future research directions.

A projection-domain iterative algorithm for metal artifact reduction by minimizing the total-variation norm and the negative-pixel energy

Metal objects in X-ray computed tomography can cause severe artifacts. The state-of-the-art metal artifact reduction methods are in the sinogram inpainting category and are iterative methods. This paper proposes a projection-domain algorithm to reduce the metal artifacts. In this algorithm, the unknowns are the metal-affected projections, while the objective function is set up in the image domain. The data fidelity term is not utilized in the objective function. The objective function of the proposed algorithm consists of two terms: the total variation of the metal-removed image and the energy of the negative-valued pixels in the image. After the metal-affected projections are modified, the final image is reconstructed via the filtered backprojection algorithm. The feasibility of the proposed algorithm has been verified by real experimental data.

Unsupervised Image-Generation Enhanced Adaptation for Object Detection in Thermal Images

Object detection in thermal images is an important computer vision task and has many applications such as unmanned vehicles, robotics, surveillance, and night vision. Deep learning-based detectors have achieved major progress, which usually need large amount of labelled training data. However, labelled data for object detection in thermal images is scarce and expensive to collect. How to take advantage of the large number labelled visible images and adapt them into thermal image domain is expected to solve. This paper proposes an unsupervised image-generation enhanced adaptation method for object detection in thermal images. To reduce the gap between visible domain and thermal domain, the proposed method manages to generate simulated fake thermal images that are similar to the target images and preserves the annotation information of the visible source domain. The image generation includes a CycleGAN-based image-to-image translation and an intensity inversion transformation. Generated fake thermal images are used as renewed source domain, and then the off-the-shelf domain adaptive faster RCNN is utilized to reduce the gap between the generated intermediate domain and the thermal target domain. Experiments demonstrate the effectiveness and superiority of the proposed method.

Deep Transfer Learning Based Human Activity Recognition By Transforming IMU Data To Image Domain Using Novel Activity Image Creation Method

Human Activity Recognition is the most popular research area in the pervasive computing field in recent years. Sensor data plays a vital role in identifying several human actions. Convolutional Neural Networks (CNNs) have now become the most recent technique in the computer vision phenomenon, but still it is premature to use CNN for sensor data, particularly in ubiquitous and wearable computing. In this paper, we have proposed the idea of transforming the raw accelerometer and gyroscope sensor data to the visual domain by using our novel activity image creation method (NAICM). Pre-trained CNN (AlexNet) has been used on the converted image domain information. The proposed method is evaluated on several online available human activity recognition dataset. The results show that the proposed novel activity image creation method (NAICM) has successfully created the activity images with a classification accuracy of 98.36% using pre trained CNN.

Degradation-Aware Deep Learning Framework for Sparse-View CT Reconstruction

Sparse-view CT reconstruction is a fundamental task in computed tomography to overcome undesired artifacts and recover the details of textual structure in degraded CT images. Recently, many deep learning-based networks have achieved desirable performances compared to iterative reconstruction algorithms. However, the performance of these methods may severely deteriorate when the degradation strength of the test image is not consistent with that of the training dataset. In addition, these methods do not pay enough attention to the characteristics of different degradation levels, so solely extending the training dataset with multiple degraded images is also not effective. Although training plentiful models in terms of each degradation level can mitigate this problem, extensive parameter storage is involved. Accordingly, in this paper, we focused on sparse-view CT reconstruction for multiple degradation levels. We propose a single degradation-aware deep learning framework to predict clear CT images by understanding the disparity of degradation in both the frequency domain and image domain. The dual-domain procedure can perform particular operations at different degradation levels in frequency component recovery and spatial details reconstruction. The peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and visual results demonstrate that our method outperformed the classical deep learning-based reconstruction methods in terms of effectiveness and scalability.