Fine-grained sketch-based image retrieval (FG-SBIR) aims at aligning images and sketches at the instance level. It is a challenging task as there are significant differences between sketch and image. Existing methods usually produce less desired performance due to the lack of large-scale fine-grained image-sketch datasets and the strong dependence on the classification models pretrained on ImageNet. In this paper, we propose a better self-supervised pre-trained FG-SBIR model which does not depend on large-scale annotated datasets. Only images and their corresponding edge maps are used at the pre-training stage. Mixed modal transformation is designed to generate different mixed-up views. The FG-SBIR model is pre-trained by minimizing the distance between the views of the same instance and then fine-tuned by a simple triplet loss. With a plain downstream network, it achieves generally better performance than state-of-the-art models on three widely used FG-SBIR datasets.

Zero-Shot Sketch-based Image Retrieval (ZS-SBIR) has been proposed recently, putting the traditional Sketch-based Image Retrieval (SBIR) under the setting of zero-shot learning. Dealing with both the challenges in SBIR and zero-shot learning makes it become a more difficult task. Previous works mainly focus on utilizing one kind of information, i.e., the visual information or the semantic information. In this paper, we propose a SketchGCN model utilizing the graph convolution network, which simultaneously considers both the visual information and the semantic information. Thus, our model can effectively narrow the domain gap and transfer the knowledge. Furthermore, we generate the semantic information from the visual information using a Conditional Variational Autoencoder rather than only map them back from the visual space to the semantic space, which enhances the generalization ability of our model. Besides, feature loss, classification loss, and semantic loss are introduced to optimize our proposed SketchGCN model. Our model gets a good performance on the challenging Sketchy and TU-Berlin datasets.

Summary Return‐Oriented Programming (ROP) attacks become very popular in recent years as these attacks can bypass traditional defense mechanisms such as data execution prevention (DEP) effectively. Previous solutions suffer from limitations in that: 1) some methods need to modify the target programs; 2) some methods introduce considerable performance cost; 3) some methods rely on the special hardware; and 4) ,most of existing methods could not provide an online protection for the target processes. In this paper, we present OnRop, an on‐the‐fly ROP attack protection system by using the commodity hardware features and OS internal facilities. Our system is compatible with the existing programs, and its protection layer can be added on demand. The experiments show that OnRop can detect ROP attacks effectively with moderate performance cost.

The online social network(OSN) phenomenon has emerged over the past ten years. Unfortunately, worms that target OSNs have become a major security threat to both the websites and their users in recent years. From our observation, we find that propagation features of these worms are highly related to the characteristics of OSNs. In this paper, we propose a new approach to do worm containment in OSNs. Firstly, we capture user connection characteristics and then establish the weighted user relationship graph which can be used in worm containment. Secondly, we propose a novel hierarchical partition algorithm to divide the nodes into difference areas based on analyzing the characteristic of the OSN structure. Finally, we leverage a new worm containment method by taking advantage of the impact of users' crucial influence on propagation. We set up experiments to verify our approach. The experiments shows that our containment strategy can restrain the worm propagation and fix the infection effectively.