Generates realistic videos from text prompts, supports arbitrary long videos, and outperforms per-frame baselines in terms of spatio-temporal quality and number of tokens per video.
Generates realistic videos from text prompts, allowing for creative and dynamic video content.
Supports generating videos of arbitrary length, making it suitable for various applications.
Outperforms per-frame baselines in terms of spatio-temporal quality, ensuring high-quality video output.
Uses a bidirectional masked transformer to generate video tokens from text, enabling efficient and effective video generation.
Employs a video encoder-decoder architecture to compress and decompress video data, reducing computational costs.
Generate videos for social media and marketing campaigns.
Create educational videos for online courses and tutorials.
Produce videos for entertainment, such as music videos and short films.
Use Phenaki for video editing and post-production tasks.
Provide text prompts to Phenaki.
Configure Phenaki's parameters and settings.
Run Phenaki to generate the video.
Edit and refine the generated video as needed.