All Projects → rystylee → Pix2pix Next Frame Prediction

rystylee / Pix2pix Next Frame Prediction

Licence: mit
pix2pix-Next-Frame-Prediction generates video by recursively generating images with pix2pix.

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Pix2pix Next Frame Prediction

tk-fangsong-font
剔骨仿宋: Experimental Fang Song style Chinese font
Stars: ✭ 93 (+481.25%)
Mutual labels:  pix2pix
pix2pix-tensorflow
A minimal tensorflow implementation of pix2pix (Image-to-Image Translation with Conditional Adversarial Nets - https://phillipi.github.io/pix2pix/).
Stars: ✭ 22 (+37.5%)
Mutual labels:  pix2pix
Igan
Interactive Image Generation via Generative Adversarial Networks
Stars: ✭ 3,845 (+23931.25%)
Mutual labels:  pix2pix
tiny-pix2pix
Redesigning the Pix2Pix model for small datasets with fewer parameters and different PatchGAN architecture
Stars: ✭ 21 (+31.25%)
Mutual labels:  pix2pix
BicycleGAN
Tensorflow implementation of the NIPS paper "Toward Multimodal Image-to-Image Translation"
Stars: ✭ 30 (+87.5%)
Mutual labels:  pix2pix
cgan-face-generator
Face generator from sketches using cGAN (pix2pix) model
Stars: ✭ 52 (+225%)
Mutual labels:  pix2pix
Pix2Pix
Image to Image Translation using Conditional GANs (Pix2Pix) implemented using Tensorflow 2.0
Stars: ✭ 29 (+81.25%)
Mutual labels:  pix2pix
Pix2pixhd
Synthesizing and manipulating 2048x1024 images with conditional GANs
Stars: ✭ 5,553 (+34606.25%)
Mutual labels:  pix2pix
pix2pix
PyTorch implementation of Image-to-Image Translation with Conditional Adversarial Nets (pix2pix)
Stars: ✭ 36 (+125%)
Mutual labels:  pix2pix
Pix2depth
DEPRECATED: Depth Map Estimation from Monocular Images
Stars: ✭ 293 (+1731.25%)
Mutual labels:  pix2pix
lin-im2im
Linear image-to-image translation
Stars: ✭ 39 (+143.75%)
Mutual labels:  pix2pix
edge2view
This is a pix2pix demo that learns from edge and translates this into view. A interactive application is also provided that translates edge to view.
Stars: ✭ 22 (+37.5%)
Mutual labels:  pix2pix
traiNNer
traiNNer: Deep learning framework for image and video super-resolution, restoration and image-to-image translation, for training and testing.
Stars: ✭ 130 (+712.5%)
Mutual labels:  pix2pix
Pix2Pix-Keras
基于pix2pix模型的动漫图片自动上色(keras实现) 2019-2-25
Stars: ✭ 95 (+493.75%)
Mutual labels:  pix2pix
Deepnude An Image To Image Technology
DeepNude's algorithm and general image generation theory and practice research, including pix2pix, CycleGAN, UGATIT, DCGAN, SinGAN, ALAE, mGANprior, StarGAN-v2 and VAE models (TensorFlow2 implementation). DeepNude的算法以及通用生成对抗网络(GAN,Generative Adversarial Network)图像生成的理论与实践研究。
Stars: ✭ 4,029 (+25081.25%)
Mutual labels:  pix2pix
coursera-gan-specialization
Programming assignments and quizzes from all courses within the GANs specialization offered by deeplearning.ai
Stars: ✭ 277 (+1631.25%)
Mutual labels:  pix2pix
Voice-Denoising-AN
A Conditional Generative Adverserial Network (cGAN) was adapted for the task of source de-noising of noisy voice auditory images. The base architecture is adapted from Pix2Pix.
Stars: ✭ 42 (+162.5%)
Mutual labels:  pix2pix
Gan Compression
[CVPR 2020] GAN Compression: Efficient Architectures for Interactive Conditional GANs
Stars: ✭ 800 (+4900%)
Mutual labels:  pix2pix
Von
[NeurIPS 2018] Visual Object Networks: Image Generation with Disentangled 3D Representation.
Stars: ✭ 497 (+3006.25%)
Mutual labels:  pix2pix
Ngx
Ngx - Neural network based visual generator and mixer
Stars: ✭ 277 (+1631.25%)
Mutual labels:  pix2pix

pix2pix-Next-Frame-Prediction

Description

pix2pix-Next-Frame-Prediction generates video by recursively generating images with pix2pix.
We use the memo's script for training and prediction.

training : affinelayer/pix2pix-tensorflow
prediction : memo/webcam-pix2pix-tensorflow

ToDo

  • [x] simple training
  • [ ] re-training using this technique
  • [ ] run realtime

Demo

Source : https://vimeo.com/223771736

Dependence

  • ffmpeg

Usage

Create Dataset

  1. Split movie to image sequence
python create_dataset.py data/XXX.mp4 data/XXX 30 png
  1. Resize source images
python pix2pix/tools/process.py --input_dir data/XXX --operation resize --output_dir data/resized
  1. Copy the file at step2
cp -r data/resized data/resized2
  1. Delete image_0000.png of resized2 and image_xxxx.png(last image) of resized1

  2. Rename the images

python rename.py data/resized
python rename.py data/resized2
  1. Combine resized 2 set of images
python pix2pix/tools/process.py --input_dir data/resized --b_dir data/resized2 --operation combine --output_dir data/combined
  • Split into train/val set
python pix2pix/tools/split.py --dir data/combined

Training

python pix2pix/pix2pix.py --mode train --output_dir dst_path --max_epochs 200 --input_dir data_path --which_direction AtoB

Generate image sequence

python generate.py models/XXX.json generate/src generate/dst num_itr

Generate movie

  1. combine
ffmpeg -r 30 -i generate/dst/%04d.png -vcodec libx264 -pix_fmt yuv420p -r 30 movie.mp4
  1. Resize
ffmpeg -i movie.mp4 -vf scale=640:-1 m.mp4

Reference

Licence

MIT

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].