Systems and methods for using redundant pictures and slice groups to encode spatially scalable H.264 Baseline profile conformant video and to route that video to endpoints of varying capabilities without using the Scalable Video extension of H.264 or transcoding. Reduced resolution versions of primary coded pictures are encoded as slice groups in a full-resolution composite pictures, which are added to the video bitstream as redundant pictures. A router then processes the spatially scaled video bitstream into separate streams having different resolutions and routes these to endpoints of varying capabilities.