Diffuse 2.0
Search documents
图生视频新玩法刷爆外网:图上画两笔就能动起来,终于告别文本提示
机器之心· 2025-08-19 02:43
Core Viewpoint - Higgsfield AI is rapidly innovating in the field of AI video generation, introducing new features that enhance user experience and creativity, while also attracting attention from major companies like Meta [3][5]. Group 1: Company Overview - Higgsfield AI was founded in October 2023 and is headquartered in San Francisco, focusing on personalized AI video tools for content creators [36]. - The company has a team size of approximately 11-50 people and raised $8 million in seed funding led by Menlo Ventures in April 2024 [36]. - Alex Mashrabov, the founder and CEO, previously worked at Snap as the head of generative AI, where he developed various AI-driven features [36][39]. Group 2: Product Features and Innovations - Higgsfield AI has launched several innovative features, including Draw-to-Video, which allows users to create videos from static images by drawing elements and inputting action commands [5][21]. - The Product-to-Video feature enables users to generate advertisement videos by simply dragging product images and adding text [30]. - The company frequently releases new functionalities, such as the UPSCALE feature for enhancing image resolution to 4K or 8K, and the STEAL feature for applying others' performance styles to virtual characters [9][10]. Group 3: User Engagement and Market Response - The Draw-to-Video feature gained over 5.3 million views on X within four days of its release, indicating strong user interest [5]. - The Product-to-Video feature has also received significant engagement, with 1.6 million views on X [6]. - Users have praised the ease of use and the ability to create high-quality video content without the need for extensive resources [43].