We are General Blockchain Inc, a company developing technology to support the AI industry using blockchain technology.
Our vision is to build a programmable, human based, artificial intelligence.
The first application of this human computer is to provide image annotation services to the machine learning and artificial intelligence community.
Please access our Image Annotation AI services here, today.
Our first product is building an AI ecosystem which collects, annotates, and trains machine learning models to recognize any object in images or other data.
Microwork is our globally distributed task system where anyone with a smart phone can earn money with simple tasks like taking a photo. Users find tasks and earn Ether, which can then be converted to the local currency. An example task would be, say, take photos of dogs. These images are then given to our image annotation team which distinguishes what is in the photo so that we can build a strong dataset. In this way we’ve created “in the wild” data, real data found in the world, as a basis for training our machine learning models.
Providing image annotation services to the Artificial Intelligence industry since 2015. Our image annotation platform, our tools, and our talented cloud workers with years of experience ensure your image annotation projects are efficiently implemented at scale. Choosing the right annotation to invest in at the right time is critical to quickly train models at low cost. Combine expensive domain experts such as yourself with talented and experienced cloud workers, working in multi-step workflows, to get the highest quality annotations and the most affordable cost.
Every pixel in an image is painstakingly masked by our in house team of talented human cloud workers. Combine their skills with your knowledge to make the most valuable data.
Masks are highly accurate and beautiful and are made by skilled professionals. Bounding Boxes can be upgraded to Masks only if required, make only the Masks you need.
Bounding Boxes are quick and cheap to make, especially with experience. Labelling boxes sometimes requires domain expertise, we have experts in house, or bring your own.
Dataset Studio creates large and diverse annotated image datasets in order to accelerate AI development, machine learning, and business intelligence. We created Dataset Studio in order to show our datasets live in a web browser to our clients, customers, and partners. We created ways to filter by dataset, classes, and attributes, which are facets of images. For example, a dataset can be fashion, a class would be dresses, and attribute might be the type of dress, hem line, or style. Dataset Studio is live, and you can view our annotated datasets in your browser now.
Using our advanced, hybrid human-computer intelligence services ecosystem, we combine all of our assets to create superior, accurate machine learning models. We can demonstrate and visualize the specific accuracies and learnings of the model as it is trained, in our web console. You don’t need to worry about how to train a model because we can do it for you. You can hire us to do the entire process or just the part you need. Feel free to contact us for a free quote.
We are excited to be able to contribute our human-machine artifical intelligence system to helping understand and diagnose the current global pandemic, the coronavirus, COVID-19. We are currently using human annotation to build a strong dataset of X-Rays and CT Scans that can be used to create highly accurate machine learning models.
Fashion is an interesting and excellent example of annotating images for machine learning. Our human annotaters draw bounding boxes around the objects in question, and also describe their type. We create classes and attributes out of these descriptions, including things like style, color, and material. These attributes are things that humans can easily notice but machines cannot yet. So, we create the opportunity for superior AI with fashion as well.
Our logo datasets are perfect for retail tasks like managing inventory and price checking. You can train models to recognize logos on store shelves, or streamline inventory management processes. Using the logo datasets, you can also create AI programs to automate inventory tracking based on the logos of thousands of different brands, and track distribution of products on shelves, check for shelf gaps, help customers find items, and more.
We’ve also collected a huge variety of images in the agricultural space, resulting in a rich dataset. Domain experts in agricultural fields are able to recognise the most challenging objects in our images such as insects, diseases and visually similar species. With us, you can manage your own Ag-Tech datasets and annotate them yourself or outsource the work to our domain experts and experienced project managers.
Our Founder, Andy Gough, started the project back in 2014 at The Blockchain University, Mountain View, California. He was thinking a lot about the future of work and how bitcoin could impact it, when he discovered ethereum and realized that smart contracts could be used to incentivize a motivated group to behave according to a pre-determined deign.
With the right interface and incentive system, motivated people could answer questions, or solve problems that no other computer could solve. By making these human intelligences available in a similar way to a machine application interface, the human and their intelligence can provide computations. This human computer is particularly good at training AI to recognize things, precisely because humans can solve these problems easily and machines cannot.
Today, the human intellectual services provided by our platform supports our products at each stage of AI development, from collecting data with our mobile apps, annotating data with our hybrid-ai or human workflows, to crafting important datasets available to the entire blockchain community, and finally to model training services and consulting.
Presentation from Blockchain University