分类： 计算机科学 >> 计算机应用技术 提交时间： 2024-08-05
摘要： The recent wave of foundation models has witnessed tremendous success in computer vision (CV) and beyond, with the segment anything model (SAM) having sparked a passion for exploring task-agnostic visual foundation models. Empowered by its remarkable zero-shot generalization, SAM is currently challenging numerous traditional paradigms in CV, delivering extraordinary performance not only in various image segmentation and multi-modal segmentation (e.g., text-to-mask) tasks, but also in the video domain. Additionally, the latest released SAM 2 is once again sparking research enthusiasm in the realm of promptable visual segmentation for both images and videos. However, existing surveys mainly focus on SAM in various image processing tasks, a comprehensive and in-depth review in the video domain is notably absent. To address this gap, this work conducts a systematic review on SAM for videos in the era of foundation models. As the first to review the progress of SAM for videos, this work focuses on its applications to various tasks by discussing its recent advances, and innovation opportunities of developing foundation models on broad applications. We begin with a brief introduction to the background of SAM and video-related research domains. Subsequently, we present a systematic taxonomy that categorizes existing methods into three key areas: video understanding, video generation, and video editing, analyzing and summarizing their advantages and limitations. Furthermore, comparative results of SAM-based and current state-of-the-art methods on representative benchmarks, as well as insightful analysis are offered. Finally, we discuss the challenges faced by current research and envision several future research directions in the field of SAM for video and beyond.
分类： 地球科学 >> 地理学 提交时间： 2023-10-17 合作期刊: 《干旱区科学》
摘要： Science and policy have been interlinked for decades and perform essential nexus conditions in the governing aspects of environmental scenarios. This review paper examines the present challenges in the sciencepolicy interface in terms of water governance in the Caspian Sea and identifies effective conditions that may be used in the current context to enhance the mechanism. The evaluation of the sciencepolicy link in the water policy of the Caspian Sea reveals a gap between knowledge producer and governance system, impeding the translation of scientific information into action. Complicated and context-dependent solutions make it challenging to establish effective sciencepolicy processes in the Caspian Sea water governance settings. Establishing a common governing authority, implementing water and resource management regulations, and protecting the natural environment through legal frameworks are crucial steps to address these concerns and ensure sustainable development. Collaboration among coastal states is essential in environmental, economic, and social aspects of regional development. However, the lack of a comprehensive approach, coherent activities, and effective utilization of national and regional power has hindered efforts to halt the environmental degradation of the Caspian Sea. Local governments need to recognize their responsibility to protect and utilize the Caspian Sea for present and future generations, considering both environmental and human security. The interlinkage of the Caspian Sea water governance with the Organization for Economic Co-operation and Development (OECD) water governance principles offers a framework for policymakers to assess gaps and make necessary amendments to existing mechanisms. Effective sciencepolicy interaction, engagement of diverse stakeholders, institutionalizing agreements, and addressing collective action issues are critical for successful water governance.
点击量
待评议
点击量
下载量
评论
