Hi Andy,
The Translate API only takes text as input. It is possible to analyze the image to extract the contained text by other means. To this purpose, you may consider using image analysis APIs, such as the Cloud Vision API, as a preliminary stage. This way one can extract text from an image through optical character recognition. Once proper text is generated from the input image, the Translate API may offer the text translation service, in a second stage.
What application do you have in mind, when mentioning the “word lens”? Please let us know if this reply addresses all your concerns.
I hope this helps. Cheers!
If you would like to perform OCR and other image operations offline, you cannot possibly use APIs that need Internet to be accessed. The Translate API and Cloud Vision API are two good options to use if your device has access to the Internet. This is possible, when traveling, even if access not always available.
If offline translation is a must, you may use other tools, offered by different vendors.