Deep learning has enabled rapid advancements in the field of image processing. Learning based approaches have achieved stunning success over their traditional signal processing-based counterparts for a variety of applications such as object detection, semantic segmentation etc. This has resulted in the parallel development of hardware architectures capable of optimizing the inferencing of deep learning algorithms in real time. Embedded devices tend to have hard constraints on internal memory space and must rely on larger (but relatively very slow) DDR memory to store vast data generated while processing the deep learning algorithms. Thus, associated systems have to be evolved to make use of the optimized hardware balancing compute times with data operations. We propose such a generalized framework that can, given a set of compute elements and memory arrangement, devise an efficient method for processing of multidimensional data to optimize inference time of deep learning algorithms for vision applications.
This paper investigates the relationship between image quality and computer vision performance. Two image quality metrics, as defined in the IEEE P2020 draft Standard for Image quality in automotive systems, are used to determine the impact of image quality on object detection. The IQ metrics used are (i) Modulation Transfer function (MTF), the most commonly utilized metric for measuring the sharpness of a camera; and (ii) Modulation and Contrast Transfer Accuracy (CTA), a newly defined, state-of-the-art metric for measuring image contrast. The results show that the MTF and CTA of an optical system are impacted by ISP tuning. Some correlation is shown to exist between MTF and object detection (OD) performance. A trend of improved AP5095 as MTF50 increases is observed in some models. Scenes with similar CTA scores can have widely varying object detection performance. For this reason, CTA is shown to be limited in its ability to predict object detection performance. Gaussian noise and edge enhancement produce similar CTA scores but different AP5095 scores. The results suggest MTF is a better predictor of ML performance than CTA.
Portraits are one of the most common use cases in photography, especially in smartphone photography. However, evaluating portrait quality in real portraits is costly, inconvenient, and difficult to reproduce. We propose a new method to evaluate a large range of detail preservation renditions on realistic mannequins. This laboratory setup can cover all commercial cameras from videoconference to high-end DSLRs. Our method is based on 1) the training of a machine learning method on a perceptual scale target 2) the usage of two different regions of interest per mannequin depending on the quality of the input portrait image 3) the merge of the two quality scales to produce the final wide range scale. On top of providing a fine-grained wide range detail preservation quality output, numerical experiments show that the proposed method is robust to noise and sharpening, unlike other commonly used methods such as the texture acutance on the Dead Leaves chart.
In this paper, we present a deep-learning approach that unifies handwriting and scene-text detection in images. Specifically, we adopt adversarial domain generalization to improve text detection across different domains and extend the conventional dice loss to provide extra training guidance. Furthermore, we build a new benchmark dataset that comprehensively captures various handwritten and scene text scenarios in images. Our extensive experimental results demonstrate the effectiveness of our approach in generalizing detection across both handwriting and scene text.
The Driver Monitoring System (DMS) presented in this work aims to enhance road safety by continuously monitoring a drivers behavior and emotional state during vehicle operation. The system utilizes computer vision and machine learning techniques to analyze the drivers face and actions, providing real-time alerts to mitigate potential hazards. The primary components of the DMS include gaze detection, emotion analysis, and phone usage detection. The system tracks the drivers eye movements to detect drowsiness and distraction through blink patterns and eye-closure durations. The DMS employs deep learning models to analyze the drivers facial expressions and extract dominant emotional states. In case of detected emotional distress, the system offers calming verbal prompts to maintain driver composure. Detected phone usage triggers visual and auditory alerts to discourage distracted driving. Integrating these features creates a comprehensive driver monitoring solution that assists in preventing accidents caused by drowsiness, distraction, and emotional instability. The systems effectiveness is demonstrated through real-time test scenarios, and its potential impact on road safety is discussed.
Optimizing exposure time for low light scenarios involves a trade-off between motion blur and signal to noise ratio. A method for defining the optimum exposure time for a given function has not been described in the literature. This paper presents the design of a simulation of motion blur and exposure time from the perspective of a real-world camera. The model incorporates characteristics of real-world cameras including the light level (quanta), shot noise and lens distortion. In our simulation, an image quality target chart called the Siemens Star chart will be used, and the simulation outputs a blurred image as if captured from a camera of set exposure and set movement speed. The resulting image is then processed in Imatest in which image quality readings will be extracted from the image and consequently the relationship between exposure time, motion blur and the image quality metrics can be evaluated.
The goal of our work is to design an automotive platform for AD/ADAS data acquisition and to demonstrate its application to behavior analysis of vulnerable road users. We present a novel data capture platform mounted on a Mercedes GLC vehicle. The car is equipped with an array of sensors and recording hardware including multiple RGB cameras, Lidar, GPS and IMU. For subsequent research on human behavior analysis in traffic scenes, we have conducted two kinds of data recordings. Firstly, we have designed a range of artificial test cases which we recorded on a safety regulated proving ground with stunt persons to capture rare events in traffic scenes in a predictable and structured way. Secondly, we have recorded data on public streets of Vienna, Austria, showing unconstrained pedestrian behavior in an urban setting, while also considering European General Data Protection Regulation (GDPR) requirements. We describe the overall framework including data acquisition and ground truth annotation, and demonstrate its applicability for the implementation and evaluation of selected deep learning models for pedestrian behavior prediction.
We have developed an assistive technology for people with vision disabilities of central field loss (CFL) and low contrast sensitivity (LCS). Our technology includes a pair of holographic AR glasses with enhanced image magnification and contrast, for example, highlighting objects, and detecting signs, and words. In contrast to prevailing AR technologies which project either mixed reality objects or virtual objects to the glasses, Our solution fuses real-time sensory information and enhances images from reality. The AR glasses technology has two advantages: it’s relatively ‘fail-safe.” If the battery dies or the processor crashes, the glasses can still function because it is transparent. The AR glasses can also be transformed into a VR or AR simulator when it overlays virtual objects such as pedestrians or vehicles onto the glasses for simulation. The real-time visual enhancement and alert information are overlaid on the transparent glasses. The visual enhancement modules include zooming, Fourier filters, contrast enhancement, and contour overlay. Our preliminary tests with low-vision patients show that the AR glass indeed improved patients' vision and mobility, for example, from 20/80 to 20/25 or 20/30.
This paper presents AInBody, a novel deep learning-based body shape measurement solution. We have devised a user-centered design that automatically tracks the progress of the body by adequately integrating various methods, including human parsing, instance segmentation, and image matting. Our system guides a user's pose when taking photos by displaying the outline of the latest picture of the user, divides the human body into several parts, and compares before and after photos of the body part level. The parsing performance has been improved through an ensemble approach and a denoising phase in our main module, Advanced Human Parser. In evaluation, the proposed method is 0.1% to 4.8% better than the other best-performing model in average precision in 3 out of 5 parts, and 1.4% and 2.4% superior in mAP and mean IoU, respectively. Furthermore, the inference time of our framework takes approximately three seconds to process one HD image, demonstrating that our structure can be applied to real-time applications.
Object detection using aerial drone imagery has received a great deal of attention in recent years. While visible light images are adequate for detecting objects in most scenarios, thermal cameras can extend the capabilities of object detection to night-time or occluded objects. As such, RGB and Infrared (IR) fusion methods for object detection are useful and important. One of the biggest challenges in applying deep learning methods to RGB/IR object detection is the lack of available training data for drone IR imagery, especially at night. In this paper, we develop several strategies for creating synthetic IR images using the AIRSim simulation engine and CycleGAN. Furthermore, we utilize an illumination-aware fusion framework to fuse RGB and IR images for object detection on the ground. We characterize and test our methods for both simulated and actual data. Our solution is implemented on an NVIDIA Jetson Xavier running on an actual drone, requiring about 28 milliseconds of processing per RGB/IR image pair.