Text this: Low-complexity single-view and multi-view visual compression for visual sensor networks using block-based compressive sensing.