feel.ai



Cloud-based Emotion Recognition in Images and Videos

feel.ai is an API cloud service for emotion recognition in images and videos. The service is based on state of the art patent-pending pattern recognition technologies developed by Neuromorphic using deep learning and biologically inspired Atificial Intelligence implemented on state of the art GPUs.
The cloud service is defined by a REST API which can be accessed using HTTPS post requests for images and videos.

Recognition of Emotion in images   Try it!

Users can post images to the API and receive instantly classification results of faces and emotions detected in the image. Images can be in any of the popular formats, including JPEG, PNG, GIF and TIFF. The feel.api will automatically detect faces in the image and classify the emotion of each face, returning the most likely emotion as well as the emotion probabilities. The location of the faces in the picture are also returned in the response.

Recognition of Emotions in Videos   Try it!

Users can post emotion recognition requests in two modes:

Batch mode: Each request posted by a user is queued for execution by the API. The user can query the status of the job, and when it is completed, the emotion recognition results are returned. Using the API facilities, the user can also inspect video analytics, including an emotion graph which shows post-processed aggregate emotion state, frame-by-frame in the video. The user has also access to media information extracted from the video which can facilitate the visulation of videos in common browsers.

Immediate mode: In this mode, the request is processed by the API immediately (i.e. the request is not queued, but handled by the API without being queued). The response time will depend on the duration of the video.