‘Eflag is a powerful AI platform that transforms any normal CCTV camera into a video analytics engine’
Syed Haseeb Arfath, Chief Technology Officer of Signpost India delineates the attributes of the company’s pioneering Eflag video analytics-based campaign exposure mapping system.
As a multi-format pan-India OOH network, Signpost India engaged with a multitude of advertising requirements that had the precise requirement of size, shape, location and duration – all curated in the hope to reach out to the exact target audience based on the unconventional assumption that the locations chosen to connect with the brand’s TG.
The Covid-19 pandemic turned the OOH industry on its head. Never before was the industry brought into direct competition with online ad platforms, as directly as it has happened now. The precise reason for OOH to take a hit on an already scarce ad budget share was the lack of transparency in ad spending ROI for understanding the exposure that was brought about for an executed ad campaign. Online ad platforms, although having a high bleed rate, provides a precise mapping of exposure with varying levels of post-campaign ad penetration analysis.
Signpost looked at this crisis as an opportunity. Online ads could never bring about the uniqueness of OOH – larger-than-life display, creative freedom, non-interruptible and the potential to turn sites for brands into iconic landmarks. If Signpost would just solve the ROI mapping, it was a revolution waiting to happen and Signpost has never shied away from a challenge of this magnitude to research and develop a deep tech solution that this demanded, leaping much ahead of the technology standards of the OOH industry. And thus Eflag was born – an artificial intelligence video analytics platform developed over 24 months and more than 50,000 man-hours of cutting-edge tech research that spans hardware, software and networking expertise.
Eflag is a powerful AI platform that transforms any normal CCTV camera into a video analytics engine. It processes the video footage from CCTV cameras mounted on top of the OOH media, facing the oncoming traffic or pedestrians and accurately detects and counts the number of vehicles or people in the CCTV footage in an almost real-time format. It further can profile the detected vehicles into the type of vehicles, their manufacturer makes name, and average speed and also reads their number plate.
In the case of pedestrians, Eflag can profile the demographics of age and gender with auto face blurring for anonymised pedestrian footfall analysis. All this enables Signpost to provide a detailed post-campaign report comprising vehicle count, vehicle type split, pedestrian count, and age and gender demography during the campaign duration. In the case of DOOH, the system precisely identifies the exposure separately for each slot time in a time-shared environment.
Now, Signpost is working on enabling a real-time dashboard for its DOOH clients that will allow brands to self-monitor the performance of their respective ad campaigns and would not have to wait for the campaigns to end for understanding their ad effectiveness. The Eflag dashboard, apart from providing live CCTV footage of the oncoming traffic and the live footage of the ad playing on each of the DOOH assets, populates in real time the exposure – precisely counting the number of vehicles passing by the media and the vehicle type split across the various time of the day. With this insight, brands could update the digital ad content on the go in between the campaign if they need to adjust the messaging based on the number of cars to 2-wheeler ratio or based on precise peak traffic times.
The Eflag Video Analytics suite is validated by the prestigious Indian Institute of Science (IISC) to be 80% accurate in 2021. Eflag has broken the 85% plus barrier and is among the world’s most accurate video analytics platforms and has also been tested by the traffic police of various cities in India for their traffic count and control.