https://github.com/suraj-nair-1/google-research/tree/master/hierarchical_foresight
공식 코드Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generation
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Script for Amee Marketing & Trading Company Short Video
(Duration: 45-60 seconds)
Opening Scene (0:00-0:05):
Scene 1: Organic & Natural Offerings (0:05-0:15):
Scene 2: Engineering & Innovation (0:15-0:25):
Scene 3: Waste & Resource Management (0:25-0:35):
Scene 4: Services & Global Reach (0:35-0:45):
Closing Scene (0:45-0:55):
End Frame (0:55-1:00):
Production Notes:
Perfect for social media ads or website headers! 🌱🚀
Gif credit: MAGVIT
Source: Photo-Realistic Video Prediction on Natural Videos of Largely Changing Frames
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generation
DDLP: Unsupervised Object-Centric Video Prediction with Deep Dynamic Latent Particles
Unsupervised Image Representation Learning with Deep Latent Particles
MotionRNN: A Flexible Model for Video Prediction with Spacetime-Varying Motions
Dynamical Diffusion: Learning Temporal Dynamics with Diffusion Models
PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning
Tell Me What Happened: Unifying Text-guided Video Completion via Multimodal Masked Video Generation
Spatiotemporal Tile-based Attention-guided LSTMs for Traffic Video Prediction
Disentangling Physical Dynamics from Unknown Factors for Unsupervised Video Prediction
MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation
Predictive Experience Replay for Continual Visual Control and Forecasting
Patch-based Object-centric Transformers for Efficient Video Generation
STDiff: Spatio-temporal Diffusion for Continuous Stochastic Video Prediction
Accurate Grid Keypoint Learning for Efficient Video Prediction
Reinforcement Learning with Action-Free Pre-Training from Videos
Video Prediction Transformers without Recurrence or Convolution
PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning
Video Prediction Models as Rewards for Reinforcement Learning
VisuoSpatial Foresight for Multi-Step, Multi-Task Fabric Manipulation
Self-Supervised Visual Planning with Temporal Skip Connections
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.