Make-A-Video: Text-To-Video Generation Without Text-Video Data | Paper Explained
π Find out how to get started using Weights & Biases π http://wandb.me/ai-epiphany π¨βπ©βπ§βπ¦ Join our Discord community π¨βπ©βπ§βπ¦ https://discord.gg/peBrCpheKE In this video I cover the latest text-to-video paper from Meta: "Make-A-Video: Text-To-Video Generation Without Text-Video Data". I walk you through the 3-stage approach that consists of: * Training a DALL-E 2 type of a model * Integrating temporal information and tuning on unlabeled videos * Fine-tuning the frame interpolation module. β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬ β Paper: https://arxiv.org/abs/2209.14792 β Website: https://makeavideo.studio/ β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬ βοΈ Timetable: 00:00 Intro 00:25 (sponsored) Weights & Biases 01:37 Going through the generations 06:15 High-level paper overview 10:50 Results 15:40 Limitations 16:30 Diving deep: DALL-E 2 backbone 23:35 Expanding to 3D - temporal info integration 32:39 Frame interpolation 37:24 3-stage training 41:28 Outro β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬ π° BECOME A PATREON OF THE AI EPIPHANY β€οΈ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar VeliΔkoviΔ β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬ πΌ LinkedIn - https://www.linkedin.com/in/aleksagordic/ π¦ Twitter - https://twitter.com/gordic_aleksa π¨βπ©βπ§βπ¦ Discord - https://discord.gg/peBrCpheKE πΊ YouTube - https://www.youtube.com/c/TheAIEpiphany/ π Medium - https://gordicaleksa.medium.com/ π» GitHub - https://github.com/gordicaleksa π’ AI Newsletter - https://aiepiphany.substack.com/ β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬β¬ #makeavideo #meta #texttovideo
