-
Chapter and Conference Paper
Revisiting TENT for Test-Time Adaption Semantic Segmentation and Classification Head Adjustment
Test-time adaption is very effective at solving the domain shift problem where the training data and testing data are sampled from different domains. However, most test-time adaption methods made their success...
-
Chapter and Conference Paper
Enhancing Adversarial Transferability from the Perspective of Input Loss Landscape
The transferability of adversarial examples enables the black-box attacks and poses a threat to the application of deep neural networks in real-world, which has attracted great attention in recent years. Regar...
-
Chapter and Conference Paper
Eventually-Consistent Replicated Relations and Updatable Views
Distributed systems have to live with weak consistency, such as eventual consistency, if high availability is the primary goal and network partitioning is unexceptional. Local-first applications are examples o...
-
Chapter and Conference Paper
Capturing the Lighting Inconsistency for Deepfake Detection
The rapid development and widely spread of deepfake techniques have raised severe societal concerns. Thus detecting such forgery contents has become a hot research topic. Many deepfake detection methods have b...
-
Chapter and Conference Paper
Elastic Deep Learning Using Knowledge Distillation with Heterogeneous Computing Resources
In deep neural networks, using more layers and parameters generally improves the accuracy of the models, which get bigger. Such big models have high computational complexity and big memory requirements, which ...
-
Chapter and Conference Paper
Noise Simulation-Based Deep Optical Watermarking
Digital watermarking is an important branch of information hiding, which effectively guarantees the robustness of embedded watermarks in distorted channels. To embed the watermark into the host carrier, tradit...
-
Chapter and Conference Paper
Supporting Undo and Redo for Replicated Registers in Collaborative Applications
A collaborative application supporting eventual consistency may temporarily violate global invariant. Users may make mistakes. Undo and redo are a generic tool to restore global invariant and correct mistakes....
-
Chapter and Conference Paper
HTCN: Harmonious Text Colorization Network for Visual-Textual Presentation Design
The selection of text color is a time-consuming and important aspect in the designing of visual-textual presentation layout. In this paper, we propose a novel deep neural network architecture for predicting te...
-
Chapter and Conference Paper
Talking Face Video Generation with Editable Expression
In rencent years, the convolutional neural network have been proved to be a great success in generating talking face. Existing methods have combined a single face image with speech to generate talking face vid...
-
Chapter and Conference Paper
Augmenting SQLite for Local-First Software
Local-first software aims at both the ability to work offline on local data and the ability to collaborate across multiple devices. CRDTs (conflict-free replicated data types) are abstractions for offline and ...
-
Chapter and Conference Paper
Towards More Powerful Multi-column Convolutional Network for Crowd Counting
Scale variation has always been one of the most challenging problems for crowd counting. By using multi-column convolutions with different receptive fields to deal with different scales in the scene, the multi...
-
Chapter and Conference Paper
Stereo Visual SLAM Using Bag of Point and Line Word Pairs
The traditional point-based SLAM algorithm performs poorly due to light changing, low-texture and highly similar scenes, while line segment features can better describe the structural information of the enviro...
-
Chapter and Conference Paper
Research on Cloud Computing Based on Deep Analysis to Typical Platforms
Cloud Computing, as a long-term dream of turning the computation to a public utility, has the potential to make IT industry great changed: making software more charming as a service and changing the way hardwa...
-
Chapter and Conference Paper
APFA: Asynchronous Parallel Finite Automaton for Deep Packet Inspection in Cloud Computing
Security in cloud computing is getting more and more important recently. Besides passive defense such as encryption, it is necessary to implement real-time active monitoring, detection and defense in the cloud...