https://arxiv.org/pdf/1704.08812.pdf
Automatic Real-time Background Cut for Portrait Videos
We in this paper solve the problem of high-quality automatic real-time background cut for 720p portrait videos. We first handle the background ambiguity issue in semantic segmentation by proposing a global background attenuation model. A spatial-temporal refinement network is developed to further refine the segmentation errors in each frame and ensure temporal coherence in the segmentation map. We form an end-to-end network for training and testing. Each module is designed considering efficiency and accuracy. We build a portrait dataset, which includes 8,000 images with high-quality labeled map for training and testing. To further improve the performance, we build a portrait video dataset with 50 sequences to fine-tune video segmentation. Our framework benefits many video processing applications.
Automatic Real-time Background Cut for Portrait Videos
We in this paper solve the problem of high-quality automatic real-time background cut for 720p portrait videos. We first handle the background ambiguity issue in semantic segmentation by proposing a global background attenuation model. A spatial-temporal refinement network is developed to further refine the segmentation errors in each frame and ensure temporal coherence in the segmentation map. We form an end-to-end network for training and testing. Each module is designed considering efficiency and accuracy. We build a portrait dataset, which includes 8,000 images with high-quality labeled map for training and testing. To further improve the performance, we build a portrait video dataset with 50 sequences to fine-tune video segmentation. Our framework benefits many video processing applications.
Gentleminds via @arxivabstract_bot
https://arxiv.org/pdf/1704.08812.pdf Automatic Real-time Background Cut for Portrait Videos We in this paper solve the problem of high-quality automatic real-time background cut for 720p portrait videos. We first handle the background ambiguity issue in…
Сегментация человека на видео.
* Кусочки фона на вход второй головы для помощи сети
* Пачка кадров для временного контекста
* Filter pruning ResNet-18 для 10х ускорения
#segmentation
* Кусочки фона на вход второй головы для помощи сети
* Пачка кадров для временного контекста
* Filter pruning ResNet-18 для 10х ускорения
#segmentation
Тут ребята применили обратный РЛ, чтобы узнать мотивации червей (не дождевых червей, а Caenorhabditis elegans). Ну то есть как обычно: награда -> стратегия, а тут посчитали что стратегия у этих червей и так оптимальная, поэтому интересно посмотреть что там за награды #reinforcementlearning
You are able to tell which nonsense word is Pokémon character and which is the latest craze in #bigdata technology? You just might qualify as #bigdata expert. Much success! https://docs.google.com/forms/d/e/1FAIpQLScRsfRHXPTuEXdNvUcI8DzJIU5iazqlpksWucPF0d8l2ztkkA/viewform
Google Docs
Is it Pokémon or #bigdata technology?
You is able to tell which nonsense word is Pokémon character and which is the latest craze in #bigdata technology? You just might qualify as #bigdata expert. Much success!
подоспел свежий фреймворк для #reinforcementlearning от фейсбука, типа очень гибкий и быстрый, позволяет запускать среды пачками и проч. https://github.com/facebookresearch/ELF
GitHub
GitHub - facebookresearch/ELF: An End-To-End, Lightweight and Flexible Platform for Game Research
An End-To-End, Lightweight and Flexible Platform for Game Research - facebookresearch/ELF