Search Results for "combine"

HunyuanCustom is a multimodal video customization framework by Tencent Hunyuan, aimed at generating customized videos featuring particular subjects (people, characters) under flexible conditions, while maintaining subject/identity consistency. It supports conditioning via image, audio, video, and text, and can perform subject replacement in videos, generate avatars speaking given audio, or combine multiple subject images. The architecture builds on HunyuanVideo, with added modules for identity reinforcement and modality-specific condition injection. Text-image fusion module based on LLaVA for improved multimodal understanding. Applicable to single- and multi-subject scenarios, video editing/replacement, singing avatars etc.

Implementation of Phenaki Video, which uses Mask GIT to produce text-guided videos of up to 2 minutes in length, in Pytorch. It will also combine another technique involving a token critic for potentially even better generations. A new paper suggests that instead of relying on the predicted probabilities of each token as a measure of confidence, one can train an extra critic to decide what to iteratively mask during sampling. This repository will also endeavor to allow the researcher to train on text-to-image and then text-to-video. ...

Implementation of Make-A-Video, new SOTA text to video generator from Meta AI, in Pytorch. They combine pseudo-3d convolutions (axial convolutions) and temporal attention and show much better temporal fusion. The pseudo-3d convolutions isn't a new concept. It has been explored before in other contexts, say for protein contact prediction as "dimensional hybrid residual networks". The gist of the paper comes down to, take a SOTA text-to-image model (here they use DALL-E2, but the same learning points would easily apply to Imagen), make a few minor modifications for attention across time and other ways to skimp on the compute cost, do frame interpolation correctly, get a great video model out. ...

...VideoCrafter2, an upgraded version, improves on its predecessor by enhancing motion dynamics and concept combinations, especially in low-data scenarios. Users can explore a wide range of creative possibilities, producing cinematic videos that combine artistic styles and real-world scenes.

Implementation of Recurrent Interface Network (RIN), for highly efficient generation of images and video without cascading networks, in Pytorch. The author unawaredly reinvented the induced set-attention block from the set transformers paper. They also combine this with the self-conditioning technique from the Bit Diffusion paper, specifically for the latents. The last ingredient seems to be a new noise function based around the sigmoid, which the author claims is better than cosine scheduler for larger images. The big surprise is that the generations can reach this level of fidelity. ...