VisionSDK is currently Alpha
1
Add our SDK to your project
The SDK is highly optimized. Our CNN model size is under 1MB and upto 62x faster.
2
Just pass the image
You can use JAVA or Kotlin to interface with the SDK
3
You APP in now intelligent
Analyze images and videos on the mobile in real time. No need for cloud vision APIs. 100% FREE.
For custom enterprise solution
Intelligent Camera
Detects people, animals, and vehicles in realtime and notifies the owner.
Watch DemoInstance Segmentation
Generates bounding boxes and masks each instance of an object in the image.
View DemoSearch for spoken phrases
Search for spoken phrases without the need for video description or subtitles.
Watch DemoFor custom enterprise solution
Our proprietary technology converts the video into a text corpus. By using NLP models we can answer questions like "When did this phrase appear?" or "When was this taught?".
Not just text, our technology can also search for real world objects like 'People', 'Vehicles', or even 'Animals' inside a video. Simply ask "Did a balck cat appear in the video?" and you will get your answer in no time.
We are working on to answer some of the toughest questions in computer vision; "When did the person wearing red T-shirt throw the ball?". By carefully mapping the relationships of objects inside the video we will be able to answer these question.
Get answers to your critical questions like "Does this video convey a positive message?" or "Will this video make our users happy?". By using complex computer vision models we can predict this within seconds.
VisionSDK is currently Alpha