AI video production is rapidly evolving, and one company at the forefront of this innovation is Runway, a generative AI startup based in New York City. Runway has recently updated its Gen-2 foundation model with a groundbreaking tool called Multi Motion Brush. This tool allows creators to incorporate multiple directions and types of motion into their AI video creations, a feat previously unmatched by any commercially available AI video projects. Other similar products on the market can only add motion to the entire image or a selected highlighted area.
The Power of Multi Motion Brush
The introduction of Multi Motion Brush signifies a significant advancement in the creative AI market. It grants users greater control over their AI-generated videos by enabling them to independently add motion to specific areas of their choice. For instance, users can dictate the movement of a face or determine the direction of the clouds in the sky. The process begins with uploading a still image as a prompt and using a digital brush controlled by the computer cursor to “paint” the desired motion onto the image. Through slider controls in Runway’s web interface, users can define the direction and intensity of the motion for each painted portion independently. The horizontal, vertical, and proximity sliders allow users to specify left/right, up/down, or closer/further movements.
Runway offers detailed instructions for users: “Each slider is controlled with a decimal point value with a range from -10 to +10. You can manually input numerical value, drag the text field left or right, or use the sliders. If you need to reset everything, click the ‘Clear’ button to reset everything back to 0.”
Runway’s Gen-2 Model and Additional Features
Runway’s Gen-2 model, unveiled in March 2023, expanded the capabilities of its predecessor, Gen-1, by introducing text, video, and image-based generation. Initially, the model could only generate clips up to four seconds. However, in August, Runway extended this capability to 18 seconds. Alongside this improvement, the company also introduced a “Director Mode,” empowering users to customize the direction and intensity/speed of camera movements in generated videos. Additionally, users gained the ability to choose from various video styles, including 3D cartoon, cinematic, and advertising.
This latest development places Runway in direct competition with other players in the AI-driven video generation field, such as Pika Labs and Stability AI. Pika Labs recently launched its web platform, Pika 1.0, for video generation, while Stability AI specializes in Stable Video Diffusion models. While these tools have seen enhancements over time, it is crucial to note that they may still yield imperfect outputs, such as blurred, incomplete, or inconsistent images or videos.
