Eyeware Data Recorder vs Vision Detector

Eyeware Data Recorder is an iOS app intended to be used as a data collection tool by the Eyeware team. The collected data is used to improve the eye tracking algorithms that are developed and distributed by Eyeware in their research projects and products. An example of such a product is the Eyeware Beam iOS app, which turns your iPhone into a head and eye tracking device for gamers and streamers. In this case, gamers and streamers would directly benefit from the data collected by testers using the Eyeware Data Recorder iOS app: Beam users will have a smoother and increasingly pleasant experience using their iPhone or iPad as a head and eye tracker. Eyeware Data Recorder is useful for scientific studies that will in the end increase the performance of eye tracking algorithms developed by Eyeware Tech. The app records the user’s face area while following a sequence of points appearing on the screen of the device. The user is asked to watch a set of points that will be used as ground-truth. The data, in the form of videos and non-personal metadata, will be uploaded to Eyeware’s servers, and will then be used to help improve the performance of Eyeware’s gaze tracking algorithms. All collected data is treated with the appropriate care in terms of confidentiality and security. The data is anonymized: users are asked for some personal details (name and email), but is stored on a separate server from the video data. Thus, no-one at Eyeware can have access to both the personal details and the video data. All collected data is not connected in any way to personal details of the data provider, i.e. the tester who is participating in the data collection activity. Collected data can be deleted at any moment by simply emailing us via the email address you provided for the sign up.

Vision Detector is an AI image analysis tool that supports CoreML machine learning models, multi-modal Large Language Models, and on-device Apple Intelligence. Choose the mode that best fits your use case. You can choose one of the following image input sources: - Live video by the built-in camera - Still image by the built-in camera - Photo Library - Files (iCloud, Google Drive, Dropbox, etc.) For live video inputs, continuous inference is performed on the camera feed. However, the frame rate and other parameters depend on the device. # Notification Center Vision Detector can call Notification Center for predefined conditions. Broadcasting notifications sends notifications to all Apple devices signed in with the same Apple Account. ## CoreML Mode Vision Detector locally executes CoreML models using Apple’s Vision framework. Before using this mode, prepare a machine learning model in the CoreML format (.mlmodel or .mlpackage) using Create ML. Import ML models via iCloud or AirDrop. Vision Detector supports the following types of CoreML models: - Image classification - Object detection - Object Segmentation - Style transfer Built-in Vision framework is also available for human mask and image classification. ## Vision LLM Mode Vision Detector can use multi-modal large language model (LLMs). Define what to do with the prompt as follows: - If the image is a screenshot, answer the name of the software - Call a notification when you see a cat - Count the number of persons in the image, and notify when it changes ## Apple Intelligence mode The images are completely processed on your device locally with your prompt.