An official website of the United States government
High-quality environment lighting is essential for creating immersive mobile augmented reality (AR) experiences. However, achieving visually coherent estimation for mobile AR is challenging due to several key limitations in AR device sensing capabilities, including low camera FoV and limited pixel dynamic ranges. Recent advancements in generative AI, which can generate high-quality images from different types of prompts, including texts and images, present a potential solution for high-quality lighting estimation. Still, to effectively use generative image diffusion models, we must address two key limitations of content quality and slow inference. In this work, we design and implement a generative lighting estimation system called CleAR that can produce high-quality, diverse environment maps in the format of 360° HDR images. Specifically, we design a two-step generation pipeline guided by AR environment context data to ensure the output aligns with the physical environment's visual context and color appearance. To improve the estimation robustness under different lighting conditions, we design a real-time refinement component to adjust lighting estimation results on AR devices. To train and test our generative models, we curate a large-scale environment lighting estimation dataset with diverse lighting conditions. Through a combination of quantitative and qualitative evaluations, we show that CleAR outperforms state-of-the-art lighting estimation methods on both estimation accuracy, latency, and robustness, and is rated by 31 participants as producing better renderings for most virtual objects. For example, CleAR achieves 51% to 56% accuracy improvement on virtual object renderings across objects of three distinctive types of materials and reflective properties. CleAR produces lighting estimates of comparable or better quality in just 3.2 seconds---over 110X faster than state-of-the-art methods. Moreover, CleAR supports real-time refinement of lighting estimation results, ensuring robust and timely updates for AR applications.
more »
« less
Joint image enhancement and localization framework for vehicle model recognition in the presence of non-uniform lighting conditions
Recognizing the model of a vehicle in natural scene images is an important and challenging task for real-life applications. Current methods perform well under controlled conditions, such as frontal and horizontal view-angles or under optimal lighting conditions. Nevertheless, their performance decreases significantly in an unconstrained environment, that may include extreme darkness or over illuminated conditions. Other challenges to recognition systems include input images displaying very low visual quality or considerably low exposure levels. This paper strives to improve vehicle model recognition accuracy in dark scenes by using a deep neural network model. To boost the recognition performance of vehicle models, the approach performs joint enhancement and localization of vehicles for non-uniform-lighting conditions. Experimental results on several public datasets demonstrate the generality and robustness of our framework. It improves vehicle detection rate under poor lighting conditions, localizes objects of interest, and yields better vehicle model recognition accuracy on low-quality input image data. Grants: This work is supported by the US Department of Transportation, Federal Highway Administration (FHWA), grant contract: 693JJ320C000023 Keywords—Image enhancement, vehicle model and
more »
« less
- Award ID(s):
- 1942053
- PAR ID:
- 10309914
- Editor(s):
- Agaian, Sos S.; Jassim, Sabah A.; DelMarco, Stephen P.; Asari, Vijayan K.
- Date Published:
- Journal Name:
- Joint image enhancement and localization framework for vehicle model recognition in the presence of non-uniform lighting conditions
- Volume:
- 11734
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Object detection and semantic segmentation are two of the most widely adopted deep learning algorithms in agricultural applications. One of the major sources of variability in image quality acquired in the outdoors for such tasks is changing lighting condition that can alter the appearance of the objects or the contents of the entire image. While transfer learning and data augmentation to some extent reduce the need for large amount of data to train deep neural networks, the large variety of cultivars and the lack of shared datasets in agriculture makes wide-scale field deployments difficult. In this paper, we present a high throughput robust active lighting-based camera system that generates consistent images in all lighting conditions. We detail experiments that show the consistency in images quality leading to relatively fewer images to train deep neural networks for the task of object detection. We further present results from field experiment under extreme lighting conditions where images without active lighting significantly lack to provide consistent results. The experimental results show that on average, deep nets for object detection trained on consistent data required nearly four times less data to achieve similar level of accuracy. This proposed work could potentially provide pragmatic solutions to computer vision needs in agriculture.more » « less
-
Abstract We propose a novel framework that combines state-of-the-art deep learning approaches with pre- and post-processing algorithms for particle detection in complex/heterogeneous backgrounds common in the manufacturing domain. Traditional methods, like size analyzers and those based on dilution, image processing, or deep learning, typically excel with homogeneous backgrounds. Yet, they often fall short in accurately detecting particles against the intricate and varied backgrounds characteristic of heterogeneous particle–substrate (HPS) interfaces in manufacturing. To address this, we've developed a flexible framework designed to detect particles in diverse environments and input types. Our modular framework hinges on model selection and AI-guided particle detection as its core, with preprocessing and postprocessing as integral components, creating a four-step process. This system is versatile, allowing for various preprocessing, AI model selections, and post-processing strategies. We demonstrate this with an entrainment-based particle delivery method, transferring various particles onto substrates that mimic the HPS interface. By altering particle and substrate properties (e.g., material type, size, roughness, shape) and process parameters (e.g., capillary number) during particle entrainment, we capture images under different ambient lighting conditions, introducing a range of HPS background complexities. In the preprocessing phase, we apply image enhancement and sharpening techniques to improve detection accuracy. Specifically, image enhancement adjusts the dynamic range and histogram, while sharpening increases contrast by combining the high pass filter output with the base image. We introduce an image classifier model (based on the type of heterogeneity), employing Transfer Learning with MobileNet as a Model Selector, to identify the most appropriate AI model (i.e., YOLO model) for analyzing each specific image, thereby enhancing detection accuracy across particle–substrate variations. Following image classification based on heterogeneity, the relevant YOLO model is employed for particle identification, with a distinct YOLO model generated for each heterogeneity type, improving overall classification performance. In the post-processing phase, domain knowledge is used to minimize false positives. Our analysis indicates that the AI-guided framework maintains consistent precision and recall across various HPS conditions, with the harmonic mean of these metrics comparable to those of individual AI model outcomes. This tool shows potential for advancing in-situ process monitoring across multiple manufacturing operations, including high-density powder-based 3D printing, powder metallurgy, extreme environment coatings, particle categorization, and semiconductor manufacturing.more » « less
-
Robust iris recognition performance remains a significant challenge, especially for off-angle images captured in less constrained environments. While convolutional neural networks (CNNs) have shown great promise in iris recognition, there is limited research on the effects of gaze-angle distortions on recognition performance and the development of dedicated frameworks for off angle iris recognition. This study investigates different recognition fusion strategies for left and right off-angle iris images using deep learning. A transfer learning approach leveraging the pre-trained AlexNet model is employed to classify iris images, where frontal-view iris images are used for training and off-angle images for testing. Three fusion strategies are explored: (i) a double model approach with decision-level fusion, where separate models are trained for left and right irises and their predictions are combined, (ii) a single model approach with feature-level fusion, where a unified model extracts and fuses features from both irises, and (iii) a single model approach with image-level fusion, where left and right iris images are merged at the input level. The performance of these methods is evaluated using accuracy as the primary metric to assess the model's generalization capabilities under off-angle conditions. Experimental results highlight the advantages and trade-offs of each fusion strategy, offering insights into the role of bilateral iris information in enhancing recognition performance. The findings of this study contribute to the development of more robust deep learning-based iris recognition systems capable of handling off-angle variations.more » « less
-
Underwater image enhancement is often perceived as a disadvantageous process to object detection. We propose a novel analysis of the interactions between enhancement and detection, elaborating on the potential of enhancement to improve detection. In particular, we evaluate object detection performance for each individual image rather than across the entire set to allow a direct performance comparison of each image before and after enhancement. This approach enables the generation of unique queries to identify the outperforming and underperforming enhanced images compared to the original images. To accomplish this, we first produce enhanced image sets of the original images using recent image enhancement models. Each enhanced set is then divided into two groups: (1) images that outperform or match the performance of the original images and (2) images that underperform. Subsequently, we create mixed original-enhanced sets by replacing underperforming enhanced images with their corresponding original images. Next, we conduct a detailed analysis by evaluating all generated groups for quality and detection performance attributes. Finally, we perform an overlap analysis between the generated enhanced sets to identify cases where the enhanced images of different enhancement algorithms unanimously outperform, equally perform, or underperform the original images. Our analysis reveals that, when evaluated individually, most enhanced images achieve equal or superior performance compared to their original counterparts. The proposed method uncovers variations in detection performance that are not apparent in a whole set as opposed to a per-image evaluation because the latter reveals that only a small percentage of enhanced images cause an overall negative impact on detection. We also find that over-enhancement may lead to deteriorated object detection performance. Lastly, we note that enhanced images reveal hidden objects that were not annotated due to the low visibility of the original images.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1117/12.2586036
