iVision analyzes images and tells us what's inside
Automate moderation of user-generated ad content. Flag the usual suspects like nudity, weapons, and illegal substances or customize to filter out specific off-brand content.
Info Station Image Recognition with Deep Convolutional Neural Networks
iVision
Visual Intelligence
As a brand, agency or marketer you need to see inside those images to take an informed business decision. iVision APIs help you sift through millions of images to identify trends that affect your business or to monitor streams of images to flag suspicious content like NSFW images, copyright materials, etc. that can affect your business reputation.
iVision APIs are robust to failure, and it can be scaled to process any number of images as per your requirements.
85 %
85% accuracy on AVA dataset85 %
accuracy on AVA dataset
Explore
Unique Solutions for Unique Business Problems
Whether you run an online marketplace, an e-commerce store, a content management platform, or a real-estate company, Info Station class-leading computer vision AI solutions power your business with the goal of maximizing your profits or understanding user activity.
iVision API analyzes your images and tells you what's inside of them.
For every image input, Info Station Visual Intelligence models will return a list of concepts, each with a probability score of how likely it is the concept is visible in the image.
iVision API finds the label(s) (names) of objects in the image along with its probability score. Image Recognition is beneficial to build visual search capability within our application letting us create a compelling discovery engine for your items.
iVision image recognition can identify a few thousand commonly occurring everyday objects from an image. Consider building a custom model for results unique to your business needs, contact us to get started.
Discover
Try the iVision API
Upload a photo to get a list of objects and their context in it, or scroll down to try the iVision OCR and Content-Classification
Combine Ads by Visual Similarity
Find content by visual similarity, keyword tag, or a combination of both so we can serve powerful image-based recommendations and related content in real-time.
Tag Everything, Get Organized
Info Station automatically tags all your images and video so you can quickly organize, manage, and search through your content. It’s fast, easy, and incredibly accurate.
WizVision OCR: Analyzes and automates the categorization and organization of unstructured content in English or Bengali
CNN+LSTM+CTC based OCR(Optical Character Recognition)
Fast access to meaningful content
Provides content tagging and classification that is distinctive to your organization. Users can feel confident that information is accurately organized and managed.
Filters out irrelevant content
Filters out content, such as Webpages about irrelavant content or documents that do not hold business value, to help ensure that only important content is retained, classified and bid on.
Content clustering and search
Allows you to move beyond simple keyword search to enable conceptual search that delivers a more relevant set of results.
Understand
Artificial Intelligence with a Vision
Learn
What is visual recognition?
Visual recognition is the ability to automatically understand any image or video based on visual elements and patterns. Info Station Vision API applies machine learning to image & video recognition, helping customers understand and manage their media. Our computer models are trained on a list of possible outputs (tags) to apply to any input (your content). Using machine learning, a process which enables a computer to learn from data and draw its own conclusions, our models are able to automatically identify the correct tags for any given image or video. These models are then made easily accessible through our simple API. Our core model identifies 11,000+ general concepts like objects, ideas, and emotions.
A model is a processing block that converts inputs (e.g. images) to predictions (e.g. concepts, bounding boxes, etc). When you run an input through a model, it will return the predictions that the model sees in the input. Each model is trained to recognize a unique set of outputs, such as concepts, faces, and colors. For example, if you run your input through the ‘food’ model, it will return concepts that the ‘food’ model knows about, such as specific dish names. If you run the same input through the ‘color’ model, it will return predictions about the dominant colors in your image. Info Station provides a variety of pre-trained models that you can use to make predictions. These models were built in-house and have been thoroughly tested.