OpenAI Introduces Sora: Text-to-Video Model for Minute-Long Videos
OpenAI introduces Sora, a text-to-video model capable of generating minute-long high-definition videos from text prompts, emphasizing its potential and limitations.
OpenAI Introduces Text-to-Video Generator "Sora"
OpenAI introduces Sora as a foundation for models to understand and simulate the real world, aiming towards Artificial General Intelligence.
I4VGEN: Enhancing Text-to-Video Diffusion Models with Image Guidance During Inference
I4VGEN is a novel inference pipeline that improves the quality of pre-trained text-to-video diffusion models by incorporating image information during inference, eliminating the need for additional training and addressing the non-zero terminal SNR issue.
事前学習済みモデルを活用したテキストからの動画生成における、画像を仲介とした効率的な手法：I4VGEN
本稿では、事前学習済みのテキスト-to-ビデオ拡散モデルに対し、追加学習なしに画像情報を活用して品質向上を実現する新しいビデオ拡散推論パイプライン「I4VGEN」を提案する。
이미지를 텍스트-비디오 생성을 위한 자유로운 디딤돌로 활용하는 I4VGen
I4VGen은 사전 훈련된 텍스트-비디오 확산 모델을 향상시키기 위해 이미지 정보를 활용하는 새로운 비디오 확산 추론 파이프라인으로, 추가적인 훈련 없이도 시각적 사실성과 텍스트 충실도가 향상된 비디오를 생성합니다.
CogVideoX: Generating High-Resolution, Long-Duration Videos from Text Using Diffusion Transformers and a Novel 3D VAE
CogVideoX introduces a novel approach to text-to-video generation, leveraging diffusion transformers, a 3D Variational Autoencoder (VAE), and an expert transformer to produce high-resolution, long-duration videos with coherent narratives and realistic motion.
MotionAura: A Novel Framework for Text-to-Video Generation and Sketch-Guided Video Inpainting Using Discrete Diffusion
MotionAura is a novel framework that leverages 3D vector-quantized diffusion models and spectral transformers to generate high-quality, temporally consistent videos from text prompts and guide video inpainting using sketches.
MotionAura: 이산 확산 모델을 사용한 고품질의 모션 일관성 있는 비디오 생성
MotionAura는 고품질의 시간적 일관성을 가진 비디오를 생성하기 위해 새로운 3D VAE 아키텍처와 스펙트럼 변환기를 활용한 텍스트-비디오 생성 모델입니다.
MotionAura：利用離散擴散模型生成高品質且運動一致的影片
MotionAura 是一種基於新型 3D VAE 和頻譜變換器的新型文字轉影片生成框架，能夠生成具有逼真時間一致性且與輸入文字提示一致的高品質影片。
HARIVO: A Novel Method for Text-to-Video Generation Using Frozen Text-to-Image Models
HARIVO is a novel single-stage method for generating diverse and high-quality videos from text prompts by leveraging the power of frozen, pre-trained text-to-image diffusion models and incorporating innovative architectural designs and loss functions to ensure temporal consistency.
© 2024 by Linnk AI