Abstract:
Human-centric computer vision tasks require pedestrian detection, which is difficult. Handcrafted and deep features have improved over the past decade. We review recent pedestrian detection advances. We first review single-spectral pedestrian detection using handcrafted and deep features. We present an extensive review of handcrafted features-based methods and find that large freedom degrees in shape and space perform better. We divide deep features-based approaches into CNN-only and CNN-and-handcrafted methods. Feature-enhanced, part-aware, and post-processing methods are the focus of our statistical analysis. Multi-spectral pedestrian detection has better illumination variance features than single-spectral detection. We also present related datasets, metrics, and a deep experimental analysis. This survey concludes with open issues and future directions.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
