As the distinction between the physical and the digital world begins to blur, it is increasingly useful to associate physical objects with digital data. This is commonly achieved with machine-readable barcodes, such as QR codes. However, the receivers’ channel qualities -which are dominated by camera resolution and distance from the code-differ widely. Consequently, existing barcode schemes either support the transmission of small payloads over long distances, or they support large payloads at the cost of requiring short distances or high-resolution receiver cameras.

We propose Focus, a system that does not require the explicit trade-off that previous work makes between code capacity and the receivers’ channel quality. Instead, by encoding data in the frequency domain of images, Focus enables receivers to decode as much data from a code as their channel supports: a nearby, high-resolution receiver can decode all data from a code, whereas a low- resolution receiver that is farther away can nonetheless partially decode the data. Focus also supports transmission of arbitrarily large payloads through videos containing sequences of codes.

 We evaluate Focus over a range of distances, receivers (smartphones and wearable technology), and displays (LCD screens, paper and ePaper). Our evaluation shows that decoding performance scales smoothly with the distance between code and receiver, and with the resolution of the receiver’s camera.