And the Phase 1 Winners are…

Congratulations to the following teams selected to move forward to Phase 2 of the NIST Enhancing Computer Vision for Public Safety Challenge!
The six winners of Phase 1 will each receive a cash prize award and are invited to participate in Phase 2 of the Challenge. The Phase 1 Winners are:

CalAster

Team iAI Tech-NJIT

Team IUPUI

Team LIVE

The University of Texas at Austin LIVE (Laboratory for Image & Video Engineering)

Team Ozer

Trueface

About the Challenge

Create a new line of research in computer vision to develop life-saving tools for public safety.
Watch the Challenge Overview Video and Challenge webinar recording to learn more!

blurry fireComputer vision works best with high-quality images. But for public safety officers in the field, things look differently. Computer vision can provide a clear picture of the situation, but dirt and grime, rain, light flares, or simply low-quality camera equipment can disrupt computer vision and remove that valuable tool.

The proposed solution is an algorithm that would go between the camera and the computer vision algorithm. This algorithm would assess the image or video quality. The computer vision system would use this information to deploy complex mitigation strategies, such as zooming, panning, changing the bit-rate, or pre-processing to de-noise, de-block, or remove airborne dust. The system could also make internal changes, such as choosing among several computer vision algorithm.

The Enhancing Computer Vision for Public Safety Challenge is designed to help develop this new line of research. In this challenge, PSCR anticipates solvers will create datasets with camera impairments and an innovative method to assess computer vision failure rates for those media.

NIST held an information webinar on September 10 where the challenge team presented a summary of the challenge and answered participant questions. A recording of the webinar is available above.

Challenge Details

The Enhancing Computer Vision for Public Safety Challenge is hosted by the Public Safety Communications Research (PSCR) Division of the National Institute of Standards and Technology (NIST). This challenge supports public safety missions by advancing the research capacity of computer vision algorithms and no-reference (NR) metrics that assess image quality or video quality.

The vision is robust computer vision systems. The approach is to develop algorithms that detect image and video quality problems that cause computer vision algorithms to fail. This is referred to as an NR metric for image and video quality assessment. The problem is that NR metric researcher is currently focused on human perceptions — not computer vision.

The goal of this challenge is to establish a starting point for new research on quality assessment for computer observers; identify camera quality problems that hinder computer vision; and quantify their impact. Each submission can either focus on one specific camera impairment with content suitable for various computer vision applications or include a variety of camera impairments with content suitable for one computer vision application.

See full challenge details at Challenge.gov.

Challenge Timeline

Download Timeline
Up to $240,000 in Prizes Available

Challenge Prizes

PSCR will award prizes valued up to $240,000 to the contestants.

Phase 1: Concept Paper

Up to 10 contestants will be awarded invitation to Phase 2 and $5,000 for dataset creation.

Phase 2: Dataset Submission

Up to 10 teams will be awarded $6,000 per team and $12,000 per team that provides datasets to CDVL for further R&D.

Two optional “Best in Show” prizes at $5,000 each may be awarded at the discretion of the judging panel. 1) Best Assessment Data Prize; and 2) Best Dataset Prize.

Challenge Partner

FirstNet