Welcome to CAM²
Continuous Analysis of Many CAMeras
Discover, Present, Analyze Real-Time Data from Worldwide Sources
What is CAM2?
- the Continuous Analysis of Many CAMeras
- A system for analyzing streaming data built by a team of Purdue University researchers.
Monitor Houston Flooding
- Network Cameras Application
- Monitoring Natural Disaster with real-time data.
News
Improving the Consistency of Object Detection
Dan Marrick
Our team has collected data and identified certain settings that result in unstable object detection. Click for
Code Review Workshop
Noah Curran
The software engineering team is putting on a workshop to improve code review within CAM2. The workshop, which will be held on March 23rd, will use real production code to demonstrate the importance of code review.
Dataset Bias Team
Ashley Kim
We are in the process of validating how dataset's distinctiveness impacts a model's accuracy meaning validating biases within the datasets.
Cross-referencing social media and public surveillance camera data for disaster purpose
Best paper award in conference: IEEE HST 2018 category: Humanitarian Assistance and Disaster Relief
Yezhi Shen
I am grateful to be the first freshman to lead a research team in CAM2. Our team is focused to measure trees' diameter with sensors and gather more image data at the same time. We are using cameras equipped with distance sensors to obtain information and using AI to recognize trees from those videos. We hope that we can make a change to other fields by implementing computer vision.
Caleb Tung receives Honorable Mention for CRA Outstanding Undergraduate Researcher
Caleb Tung, ECE Senior, along with Ling Zhang was honored by the Computing Research Association for their undergraduate research with CAM2. Caleb, along with the rest of the Parallel Computing for Vision subteam, is researching the capability of modern object detectors to accurately and consistently detect objects in CAM2 cameras. He will also be presenting on their research so far as lead author of a paper for the IEEE International Conference on Multimedia Information Processing and Retrieval.
Visiting Student to National University of Singapore (NUS) and Seoul National University (SNU)
Funded by an NSF Grant, Kent Gauen visited two universities over summer 2018 for one month each: NUS and SNU. At NUS, Kent helped in the writing of the NUS Human Activity Recognition (HAR) grant for wearable devices. At SNU, Kent further developed the collaboration between SNU and Purdue University.
