Related papers: Semantic GUI Scene Learning and Video Alignment fo…
Domain shift remains a persistent challenge in deep-learning-based computer vision, often requiring extensive model modifications or large labeled datasets to address. Inspired by human visual perception, which adjusts input quality through…
End-to-end autonomous driving requires adaptive and robust handling of complex and diverse traffic environments. However, prevalent single-mode planning methods attempt to learn an overall policy while struggling to acquire diversified…
The modern software development landscape has seen a shift in focus toward mobile applications as "smart" devices near ubiquitous adoption. Due to this trend, the complexity of mobile applications has been increasing, making development and…
Vision Transformers (ViTs) have outperformed traditional Convolutional Neural Network architectures and achieved state-of-the-art results in various computer vision tasks. Since ViTs are computationally expensive, the models either have to…
One of the primary mechanisms by which developers receive feedback about in-field failures of software from users is through bug reports. Unfortunately, the quality of manually written bug reports can vary widely due to the effort required…
In this paper, we introduce Janus, an autoregressive framework that unifies multimodal understanding and generation. Prior research often relies on a single visual encoder for both tasks, such as Chameleon. However, due to the differing…
Bug reports are vital for software maintenance that allow users to inform developers of the problems encountered while using software. However, it is difficult for non-technical users to write clear descriptions about the bug occurrence.…
Long video understanding (LVU) remains a core challenge in multimodal learning. Although recent vision-language models (VLMs) have made notable progress, existing benchmarks mainly focus on either fine-grained perception or coarse…
Text-to-3D generation often suffers from the Janus problem, where objects look correct from the front but collapse into duplicated or distorted geometry from other angles. We attribute this failure to viewpoint bias in 2D diffusion priors,…
Video anomaly detection (VAD) has been paid increasing attention due to its potential applications, its current dominant tasks focus on online detecting anomalies% at the frame level, which can be roughly interpreted as the binary or…
Scene graph generation (SGG) analyzes images to extract meaningful information about objects and their relationships. In the dynamic visual world, it is crucial for AI systems to continuously detect new objects and establish their…
Graph Neural Networks (GNNs) have demonstrated remarkable performance across various applications, yet they are vulnerable to sophisticated adversarial attacks, particularly node injection attacks. The success of such attacks heavily relies…
Scene text image super-resolution has significantly improved the accuracy of scene text recognition. However, many existing methods emphasize performance over efficiency and ignore the practical need for lightweight solutions in deployment…
In issue-tracking systems, incorporating screenshots significantly enhances the clarity of bug reports, facilitating more efficient communication and expediting issue resolution. However, determining when and what type of visual content to…
The HyperText Markup Language 5 (HTML5) <canvas> is useful for creating visual-centric web applications. However, unlike traditional web applications, HTML5 <canvas> applications render objects onto the <canvas> bitmap without representing…
Recent advances in vision tasks (e.g., segmentation) highly depend on the availability of large-scale real-world image annotations obtained by cumbersome human labors. Moreover, the perception performance often drops significantly for new…
Current feature matching methods focus on point-level matching, pursuing better representation learning of individual features, but lacking further understanding of the scene. This results in significant performance degradation when…
Video Anomaly Detection (VAD) aims to identify anomalous events in videos and accurately determine their time intervals. Current VAD methods mainly fall into two categories: traditional DNN-based approaches that focus on temporal…
Semantic segmentation methods have achieved outstanding performance thanks to deep learning. Nevertheless, when such algorithms are deployed to new contexts not seen during training, it is necessary to collect and label scene-specific data…
Semantic segmentation plays a crucial role in enabling comprehensive scene understanding for robotic systems. However, generating annotations is challenging, requiring labels for every pixel in an image. In scenarios like autonomous…