Monocular 3D object detection has gained significant attention due to its cost-effectiveness and practicality in real-world applications. However, existing monocular methods often struggle with depth estimation and spatial consistency, limiting their accuracy in complex environments. In this work, we introduce a Temporal Deep Feature Learning framework, which enhances monocular 3D object detection by integrating temporal features across sequential frames. Our approach leverages a novel deep feature auxiliary module based on convolutional recurrent structures, effectively capturing spatiotemporal information to improve depth perception and detection robustness. The proposed module is model-agnostic and can be seamlessly integrated into various existing monocular detection frameworks. Extensive experiments across multiple state-of-the-art monocular 3D object detection models demonstrate consistent performance improvements, particularly in detecting small or partially occluded objects. Our results highlight the effectiveness and generalizability of the proposed approach, making it a promising solution for real-world autonomous perception systems. The source code of this work is at: https://github.com/Shuray36/MonoTDF-Temporal-Deep-Feature-Learning-for-Generalizable-Monocular-3D-Object-Detection .

3D detection of vehicles is an essential component for autonomous driving applications. Nevertheless, collecting the supervised training data for learning 3D vehicle detectors would be costly (e.g. utilization of expensive LiDAR sensors) and labor-intensive (for human annotation). In comparison to 3D detection, 2D object detection has achieved a welldeveloped status, boosting stable and robust performance with widespread application in numerous fields, thanks to the large scale (i.e. amount of samples) of existing training datasets of 2D object detection. Hence, in our work, we propose to realize 3D detection via leveraging the robustness of 2D detectors and developing a network that lifts 2D detections to 3D. With the flexibility of building upon various backbone models (e.g. the models which take image regions detected by 2D detector as inputs to predict their corresponding 3D bounding boxes, or the existing monocular 3D detection models which have the intermediate output of 2 D bounding boxes), we propose several geometry-driven objectives, including projection consistency loss, geometry depth loss, and opposite bin loss, to improve the training upon 2D-to-3D lifting. Our extensive experimental results demonstrate that our proposed geometrydriven objectives not only contribute to the superior results of 3D detection but also provide better generalizability across datasets.

Having a well sleep quality is important factor in our daily life. The evaluation of sleep stages has become an important issue due to the distribution of sleep stages across a whole night relates to sleep quality. This study aims to propose a sleep classification system, consists of a preliminary wake detection rule, sleep feature extraction, fuzzy c-means based dimension reduction, support vector machine with radial basis function kernel, and adaptive adjustment scheme, with only FP1 and FP2 electroencephalography. Compared with the results from the sleep technologist, the average accuracy and Kappa coefficient of the proposed sleep classification system is 70.92% and 0.6130, respectively, for individual 10 normal subjects. Thus, the proposed sleep classification system could provide a preliminary report of sleep stages to assistant doctors to make decision if a patient needs to have a detailed testing in a sleep laboratory.

Monitoring the neurophysiological activities of driver in an operational environment poses a severe measurement challenge using a current laboratory-oriented biosensor technology. The aims of this research are to 1) introduce a dry and wireless EEG system used for conveniently recording EEG signals from forehead regions, 2) propose an effective system for processing EEG recordings and translating them into the vigilance level, and 3) implement the proposed system with a JAVA-based graphical user interface (GUI) for online analysis. To validate the performance of the proposed system, this study recruited eight voluntary subjects to participate a 90-min sustained-attention driving task in a virtual-realistic driving environment. Physiological evidence obtained from the power spectral analysis showed that the dry EEG system could distinguish an alert EEG from a drowsy EEG by evaluating the spectral dynamics of delta and alpha activities. Furthermore, the experimental result of the comparison of the prediction performance using four forehead electrode sites (AF8, FP2, FP1, and AF7) implied that a single-electrode EEG signal used in the mobile and wireless EEG system is able to obtain a high prediction accuracy (~93%). Taken together, the proposed system applied a dry-EEG device combined with an effective algorithm can be a promising technology for real driving applications.

Recent studies show that hyperspectral image classification techniques that use both spectral and spatial information are more suitable, effective, and robust than those that use only spectral information. Using a spatial-contextual term, this study modifies the decision function and constraints of a support vector machine (SVM) and proposes two kinds of spatial-contextual SVMs for hyperspectral image classification. One machine, which is based on the concept of Markov random fields (MRFs), uses the spatial information in the original space (SCSVM). The other machine uses the spatial information in the feature space (SCSVMF), i.e., the nearest neighbors in the feature space. The SCSVM is better able to classify pixels of different class labels with similar spectral values and deal with data that have no clear numerical interpretation. To evaluate the effectiveness of SCSVM, the experiments in this study compare the performances of other classifiers: an SVM, a context-sensitive semisupervised SVM, a maximum likelihood (ML) classifier, a Bayesian contextual classifier based on MRFs (ML_MRF), and nearest neighbor classifier. Experimental results show that the proposed method achieves good classification performance on famous hyperspectral images (the Indian Pine site (IPS) and the Washington, DC mall data sets). The overall classification accuracy of the hyperspectral image of the IPS data set with 16 classes is 95.5%. The kappa accuracy is up to 94.9%, and the average accuracy of each class is up to 94.2%.

In this study, a novel spatial information based support vector machine for hyperspectral image classification, named spatial-contextual semi-supervised support vector machine (SC3SVM), is proposed. This approach modifies the SVM algorithm by using the spectral information and spatial-contextual information. The concept of SC3SVM is to utilize other information, obtain from the pixels of a neighborhood system in the spatial domain, to modify the effective of each patterns. Experimental results show a sound performance of classification on the famous hyperspectral images, Indian Pine site. Especially, the overall classification accuracy of whole hyperspectral image (Indian Pine site with 16 classes) is up to 96.4%, the kappa accuracy is up to 95.9%.

Combining multiple classifiers is a natural way to discover useful information and improve the performances of individual classifiers. It's based on the combination of the outputs of an ensemble of different classifiers. When interactions exist in combining multiple classifiers, fuzzy integral would be a valid method to fuse multiple classifiers. In this fuzzy fusion approach, the fuzzy measure plays an important role. Liu proposed a novel fuzzy measure, L-measure, which is more sensitive than some common measures, like ¿-measure, P-measure and V-measure. In this paper, we would combine the multiple classifiers by Choquet integral wtih this Z-measure.

In the paper, an efficient VLSI architecture for a 8/spl times/8 two-dimensional discrete cosine transform and inverse discrete cosine transform (2-D DCT/IDCT) with a new 1-D DCT/IDCT algorithm is presented. The proposed new algorithm makes sure all coefficients are positive to simplify the design of multipliers and the coefficients have less round-off error than Lee's algorithm. For computing 2-D DCT/IDCT, the row-column decomposition method is used, and the design of 1-D DCT/IDCT requires only 9 multipliers and 21 adders/subtractors. This chip is synthesized with 0.6 /spl mu/m standard cell library and 1P3M CMOS technology, and it can be operated up to 100 MHz.