VideoIQ Rialto R SeriesBy Ethan Ace, Published on Apr 29, 2013
After years of pushing their own analytics enabled cameras, VideoIQ has opened their platform up to third party IP cameras with the Rialto R series [link no longer available]. In this note, we look at this new line, its features and integration, and compare to competitive analytics.
The Rialto R series is a chassis and blade based appliance line, supporting up to sixteen cameras (four blades) in a 1U rack mount unit with 2TB of on-board storage. In a departure from earlier products, the R series supports streams from third party IP cameras, in addition to analog cameras, via 4-port blades.
Each IP blade accepts four standard definition streams or two high definition (1080p/2MP max) streams, via RTSP. Neither direct camera integrations nor ONVIF are supported, but VideoIQ says this may be added if need arises. This allows preferred camera lines to be used with VideoIQ analytics, or specialized cameras (low light or WDR) where applicable. Integrators were previously locked in to VideoIQ's own cameras only.
The Rialto R series is compatible with all VMS platforms currently supporting VideoIQ. It may also be integrated via RTSP output of each stream. However, analytics must be configured using VideoIQ's View client, as they cannot be set up via web interface or third-party client.
Users have three options for viewing cameras connected to the Rialto:
- First, and most typical, users may view the stream from the Rialto only. This method embeds bounding boxes and sends alarm data to the VMS, so both are viewed at once. This allows an operator to correlate alarms to where it is happening on screen.
- Second, users may view the original camera stream, and receive only alarm data in the VMS, not Rialto video. This may be less usable to operators, as no bounding box or other visual data is provided.
- Third, both may be viewed simultaneously. This may be preferable if using megapixel cameras but sending lower resolution stream to the Rialto for processing. The original stream, 5MP for example, may be viewed at its full size, while a secondary 720p or 1080p stream is sent to the Rialto for processing.
The Rialto R series is expected to be available in Q2 2013. Pricing starts at $6,335 USD MSRP for an 8-channel version (chassis and two blades, either IP or analog). Based on dealer discounts, we project street pricing to be ~$5,200-5,500 for the 8-channel base unit. Additional 4-port blades are $1,330 MSRP each, or about $1,100-1,200 street price.
Performance Compared to Existing Line
In their current iCVR line, VideoIQ has full control over both hardware and software, and are able to tweak each for better analytic performance. Third party cameras with unpredictable image quality may decrease accuracy compared to their own iCVRs. VideoIQ addresses this in two ways:
- Reduced effective range: VideoIQ recommends third party cameras cover only about 60% of the effective range of their own. This means a ~200' HFOV at a range of ~250' using a third-party 1080p camera versus ~300' HFOV at 500' for an iCVR-HD. These ranges may be exceeded, but are what VideoIQ considers "safe" recommendations.
- Less detail required: VideoIQ points out that analytics are looking for shape, size, speed, and movements in objects, and do not require fine details which human operators may look for. This means that video sometimes regarded as lower quality may work fine for analytics. However, ff imaging is so poor that subjects cannot be made out against the background, detection will be negatively impacted.
Based on a street price of ~$5,500, the R series has a per channel price of $650-700. Competitive server-based products, such as AgentVI, normally start in the range of ~$600-800 per channel license, pricing the Rialto in the same range when using SD cameras. Pricing increases to ~$1,375 per channel when using HD cameras. However, HD analytics are uncommon among competitors, who normally use only 4CIF streams for analysis.
Other platforms may or may not require an additional server, increasing their cost. AgentVI, for example, runs a portion of their analytics on the camera, with some centralized processing, which can often be the VMS server. Others, such as Mate are completely server-based, requiring a dedicated server. Additionally, some of these platforms may charge separately for different behaviors (vehicle vs. human vs. object counting, etc.), instead of including all rules for as VideoIQ does, further increasing cost.
Compared to VideoIQ's own HD cameras, the Rialto R is expensive at small channel counts. Assuming a third-party outdoor 1080p camera costs ~$1,000 (such as Bosch, Axis, Sony, etc.), and an iCVR-HD sells for ~$1,900, the Rialto is more expensive until about six channels, when costs are about equal. Those looking to add 1-4 channels to a system may be better off replacing cameras with iCVRs. In systems with existing cameras, the Rialto may still be preferable, depending on where cameras are mounted, as labor costs to remove and replace cameras could make up this difference.
The Rialto R series has two potential positives compared to VideoIQ's previous line:
- First, integrators may prefer to deploy their preferred cameras with IP blades, instead of VideoIQ own iCVRs. This may be a more appealing option for those looking for analytics, especially given VideoIQ's status as one of integrator's favored analytics in our surveys.
- Accepting IP streams opens up existing systems to VideoIQ's analytics, where adding them previously meant replacing cameras with iCVRs.
However, its high initial cost may limit its use in greenfield projects, with users opting for iCVRs or other options instead.