Drag & Drop or Click to Upload
Upload any image to identify objects
AI Object Recognition — Privately in Your Browser
Upload any image to identify objects
TagDiet uses Google's MobileNet neural network running on TensorFlow.js to identify objects in your photos — all without uploading a single byte to any server. The AI model runs entirely in your web browser.
Powered by ImageNet's 1,000+ categories, TagDiet can identify:
When you upload an image, TagDiet runs it through MobileNet V2 (a lightweight convolutional neural network optimized for mobile and web). The model analyzes visual features and returns its top predictions with confidence scores.
Your images never leave your device. The AI model is downloaded once (~4MB) and cached in your browser. All inference happens locally using your device's CPU or GPU. This is the PhotoDiet promise: browser-based AI, zero data transfer.