Play 2 Live

Play2Live implements interactive tasks for streamers based on neural networks

Play2Live (P2L) – the world’s first decentralized streaming platform for gamers and e-sports fans – has implemented algorithms for real-time monitoring of video streaming, video content recognition and complex objects based on the neural network, to define interactive tasks for streamers and spectators, and to verify that the task has been completed. This will provide a new level of interaction between platform users and establish a saving for the LUC (Level Up Coin) token – the only payment method in P2L.

Play2Live users will vote with LUC tokens for different tasks and set their price for streamers. (PRNewsfoto/Play2Live)

Using an advanced task manager, viewers can define tasks for streamers by choosing different conditions. By voting on P2L with LUC tokens, users can set the price of delivering a specific task for their favorite streamers, including: challenging the streamer to complete the game at the highest difficulty level, or use a specific weapon, equipment or skill in a given period of time on a specific location (for example, playing Fortnite, eliminate 10 opponents in Flush Factory using Sniper Rifles), continue streaming for three hours in a row or start a flow in a different game, etc.

Any other user can support tasks with LUC tokens or assign their own tasks. A streamer can perform all the tasks and receive all the chips or perform some of them and receive less. If successful, the streamer receives a reward equal to the price of the task, and in case of failure all the chips are returned to the spectators. How to determine if a task has been completed by the streamer or not? The neural network monitors the flow and determines with great accuracy whether the task has been completed. In fact, the task itself is a smart contract with a deposit in LUC tokens.

Most actively engaged users receive LUC rewards. This allows the gamification of the entire platform; the more users actively participate in the various activities, the more they earn tokens.

Alexey Burdyko, CEO and founder of Play2Live  :

”  Play2Live facilitates many types of interaction between a streamer and a viewer, including a two-way system of tasks, an optimized content generation process, and many more. Compared to the simplistic gift and chat options provided by existing platforms, this is a real step forward in terms of interaction. We use a computer vision algorithm to analyze video streaming. These are neural networks formed for recognition on their own datasets, including Time Series and OCR, and HUD games. We performed an overall optimization of the algorithms to speed up the system’s running time. This will start the flow analysis even on the equipment without GPU. In the future, this will also allow scanning on tape drive equipment rather than Play2Live servers, and transmitting metadata with the video stream along with the results of the scan. We are confident that the implementation of this feature will change the way that this type of content is viewed forever.  

Vladislav Arbatov, director of technology at Play2Live, is responsible for training the neural network. Arbatov is the author and leader of AI developments in the fields of computer vision and natural language processing, the creator of efficient algorithms for automated visual recognition of the human condition, the creator of the first service Neuronaming IA and a number of dialogue systems for large companies.

Vladislav Arbatov, Head of Technology at Play2Live  :

”  Interactive tasks will be developed for each popular game, and by the end of the year, this feature will be available for more than 300 games. The internal training system of a neural network will allow to add new types of events as quickly as possible. We also plan to work closely with the user community – we will be asking fans what tasks would be most interesting for a particular game.  

Register for the MVP of the streaming platform here:





Leave A Comment