Features Business & Marketing
Ask the Expert: Why is there a performance gap between various H.264 implementations?

H.264 is used to reduce the amount of bandwidth required to transmit and store video, offering new possibilities to reduce storage costs and increase efficiency. With H.264 compression, a video system can capture every detail in an image without any compromise in frame rate – an indispensable feature for object recognition, such as license plates or a person’s face. It reduces the size of recorded video by more than 80 per cent compared with the Motion JPEG format, by as much as 50 per cent compared with the traditional MPEG-2 Part 2 standard, and by 30-50 per cent compared with MPEG-4 compression. This means that much less network bandwidth is required for the video stream, resulting in a higher video quality per given bit rate.

Despite being the latest video compression technology standard, the actual performance of H.264 varies significantly. The amount of compression and the resulting image quality are dependent upon how the technology is implemented. The H.264 standard includes several profiles, each one comprising a number of blocks. The profiles define the maximum feature set, but it is up to the manufacturer to determine the features in their own implementation of the technology. The way in which a manufacturer implements the technology can affect the video quality and bandwidth performance of the device. Two profiles – Baseline and Main – are relevant to video surveillance.

The standard Baseline profile is the minimum recommendation for H.264 video implementation. It delivers robust compression capabilities, reducing storage requirements by up to 30 per cent compared to MPEG-4. Baseline profile is generally used on cameras and encoders without hardware acceleration for H.264.

Offering up to 50 per cent greater compression efficiency compared to MPEG-4, without sacrificing image quality, the Main profile is the top-of-the-line H.264 implementation for video surveillance. As it utilizes all coding blocks, H.264 Main Profile encoding is more complex, so devices supporting Main profile need to have powerful processing performance. Main Profile H.264 compression technology requires roughly twice the processing of equivalent MPEG-4 or MJPEG video.

This greater complexity also means it takes longer to encode the video, which can translate into lag time – or a delay between when events occur and when the operator sees the events on his monitor. Called latency, it can be an issue when operating pan-tilt-zoom cameras, as too much latency may not be ideal for live viewing or camera control. If latency is a concern for an end user, be sure to choose cameras that have the ability to transmit multiple independent streams. For example, some IP video devices are able to transmit multiple streams at different rates and with different compression. The device could simultaneously generate two independent H.264 streams with up to 4CIF resolution at 25/30 (PAL/NTSC), images per second; one stream could be a low latency H.264 Baseline Profile stream for live viewing and pan-tilt-zoom control, while the other is a H.264 Main Profile stream for recording high-quality images with maximum compression.

While there are differences in how manufacturers apply H.264 compression technology, the overall benefits of it – despite which profile is implemented – are clear. It delivers lower storage costs without loss of image quality, frame rate and retention time. And, it gives end users higher quality images and higher frame rates at the same drive size, retention time and bandwidth.

Cheryl Bard is a product marketing manager for Bosch Security Systems, Inc. responsible for educating the market on advanced surveillance technologies that improve image quality for maximum usability. She can be reached at cheryl.bard@us.bosch.com.