Influence of Camera-LiDAR Configuration on 3D Object Detection for Autonomous Driving

Cameras and LiDARs are both important sensors for autonomous driving, playing critical roles for 3D object detection. Camera-LiDAR Fusion has been a prevalent solution for robust and accurate autonomous driving perception. In contrast to the vast majority of existing arts that focus on how to improve the performance of 3D target detection through cross-modal schemes, deep learning algorithms, and training tricks, we devote attention to the impact of sensor configurations on the performance of learning-based methods. To achieve this, we propose a unified information-theoretic surrogate metric for camera and LiDAR evaluation based on the proposed sensor perception model. We also design an accelerated high-quality framework for data acquisition, model training, and performance evaluation that functions with the CARLA simulator. To show the correlation between detection performance and our surrogate metrics, We conduct experiments using several camera-LiDAR placements and parameters inspired by self-driving companies and research institutions. Extensive experimental results of representative algorithms on NuScenes dataset validate the effectiveness of our surrogate metric, demonstrating that sensor configurations significantly impact point-cloud-image fusion based detection models, which contribute up to 30% discrepancy in terms of average precision.

AVM-SLAM: Semantic Visual SLAM with Multi-Sensor Fusion in a Bird's Eye View for Automated Valet Parking

Automated Valet Parking (AVP) requires precise localization in challenging garage conditions, including poor lighting, sparse textures, repetitive structures, dynamic scenes, and the absence of Global Positioning System (GPS) signals, which often pose problems for conventional localization methods. To address these adversities, we present AVM-SLAM, a semantic visual SLAM framework with multi-sensor fusion in a Bird's Eye View (BEV). Our framework integrates four fisheye cameras, four wheel encoders, and an Inertial Measurement Unit (IMU). The fisheye cameras form an Around View Monitor (AVM) subsystem, generating BEV images. Convolutional Neural Networks (CNNs) extract semantic features from these images, aiding in mapping and localization tasks. These semantic features provide long-term stability and perspective invariance, effectively mitigating environmental challenges. Additionally, data fusion from wheel encoders and IMU enhances system robustness by improving motion estimation and reducing drift. To validate AVM-SLAM's efficacy and robustness, we provide a large-scale, high-resolution underground garage dataset, available at https://github.com/yale-cv/avm-slam. This dataset enables researchers to further explore and assess AVM-SLAM in similar environments.

UnitModule: A Lightweight Joint Image Enhancement Module for Underwater Object Detection

Underwater object detection faces the problem of underwater image degradation, which affects the performance of the detector. Underwater object detection methods based on noise reduction and image enhancement usually do not provide images preferred by the detector or require additional datasets. In this paper, we propose a plug-and-play Underwater joint image enhancement Module (UnitModule) that provides the input image preferred by the detector. We design an unsupervised learning loss for the joint training of UnitModule with the detector without additional datasets to improve the interaction between UnitModule and the detector. Furthermore, a color cast predictor with the assisting color cast loss and a data augmentation called Underwater Color Random Transfer (UCRT) are designed to improve the performance of UnitModule on underwater images with different color casts. Extensive experiments are conducted on DUO for different object detection models, where UnitModule achieves the highest performance improvement of 2.6 AP for YOLOv5-S and gains the improvement of 3.3 AP on the brand-new test set (URPCtest). And UnitModule significantly improves the performance of all object detection models we test, especially for models with a small number of parameters. In addition, UnitModule with a small number of parameters of 31K has little effect on the inference speed of the original object detection model. Our quantitative and visual analysis also demonstrates the effectiveness of UnitModule in enhancing the input image and improving the perception ability of the detector for object features.

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

Seasonal-trend decomposition is one of the most fundamental concepts in time series analysis that supports various downstream tasks, including time series anomaly detection and forecasting. However, existing decomposition methods rely on batch processing with a time complexity of O(W), where W is the number of data points within a time window. Therefore, they cannot always efficiently support real-time analysis that demands low processing delay. To address this challenge, we propose OneShotSTL, an efficient and accurate algorithm that can decompose time series online with an update time complexity of O(1). OneShotSTL is more than $1,000$ times faster than the batch methods, with accuracy comparable to the best counterparts. Extensive experiments on real-world benchmark datasets for downstream time series anomaly detection and forecasting tasks demonstrate that OneShotSTL is from 10 to over 1,000 times faster than the state-of-the-art methods, while still providing comparable or even better accuracy.

DAA: A Delta Age AdaIN operation for age estimation via binary code transformer

Naked eye recognition of age is usually based on comparison with the age of others. However, this idea is ignored by computer tasks because it is difficult to obtain representative contrast images of each age. Inspired by the transfer learning, we designed the Delta Age AdaIN (DAA) operation to obtain the feature difference with each age, which obtains the style map of each age through the learned values representing the mean and standard deviation. We let the input of transfer learning as the binary code of age natural number to obtain continuous age feature information. The learned two groups of values in Binary code mapping are corresponding to the mean and standard deviation of the comparison ages. In summary, our method consists of four parts: FaceEncoder, DAA operation, Binary code mapping, and AgeDecoder modules. After getting the delta age via AgeDecoder, we take the average value of all comparison ages and delta ages as the predicted age. Compared with state-of-the-art methods, our method achieves better performance with fewer parameters on multiple facial age datasets.

Implicit Stochastic Gradient Descent for Training Physics-informed Neural Networks

Physics-informed neural networks (PINNs) have effectively been demonstrated in solving forward and inverse differential equation problems, but they are still trapped in training failures when the target functions to be approximated exhibit high-frequency or multi-scale features. In this paper, we propose to employ implicit stochastic gradient descent (ISGD) method to train PINNs for improving the stability of training process. We heuristically analyze how ISGD overcome stiffness in the gradient flow dynamics of PINNs, especially for problems with multi-scale solutions. We theoretically prove that for two-layer fully connected neural networks with large hidden nodes, randomly initialized ISGD converges to a globally optimal solution for the quadratic loss function. Empirical results demonstrate that ISGD works well in practice and compares favorably to other gradient-based optimization methods such as SGD and Adam, while can also effectively address the numerical stiffness in training dynamics via gradient descent.

Interactive Log Parsing via Light-weight User Feedbacks

Template mining is one of the foundational tasks to support log analysis, which supports the diagnosis and troubleshooting of large scale Web applications. This paper develops a human-in-the-loop template mining framework to support interactive log analysis, which is highly desirable in real-world diagnosis or troubleshooting of Web applications but yet previous template mining algorithms fails to support it. We formulate three types of light-weight user feedbacks and based on them we design three atomic human-in-the-loop template mining algorithms. We derive mild conditions under which the outputs of our proposed algorithms are provably correct. We also derive upper bounds on the computational complexity and query complexity of each algorithm. We demonstrate the versatility of our proposed algorithms by combining them to improve the template mining accuracy of five representative algorithms over sixteen widely used benchmark datasets.

Motion-based Post-Processing: Using Kalman Filter to Exclude Similar Targets in Underwater Object Tracking

Visual tracker includes network and post-processing. Despite the color distortion and low contrast of underwater images, advanced trackers can still be very competitive in underwater object tracking because deep learning empowers the networks to discriminate the appearance features of the target. However, underwater object tracking also faces another problem. Underwater targets such as fish and dolphins, usually appear in groups, and creatures of the same species usually have similar expressions of appearance features, so it is challenging to distinguish the weak differences characteristics only by the network itself. The existing detection-based post-processing only reflects the results of single frame detection, but cannot locate real targets among similar targets. In this paper, we propose a new post-processing strategy based on motion, which uses Kalman filter (KF) to maintain the motion information of the target and exclude similar targets around. Specifically, we use the KF predicted box and the candidate boxes in the response map and their confidence to calculate the candidate location score to find the real target. Our method does not change the network structure, nor does it perform additional training for the tracker. It can be quickly applied to other tracking fields with similar target problem. We improved SOTA trackers based on our method, and proved the effectiveness of our method on UOT100 and UTB180. The AUC of our method for OSTrack on similar subsequences is improved by more than 3% on average, and the precision and normalization precision are improved by more than 3.5% on average. It has been proved that our method has good compatibility in dealing with similar target problems and can enhance performance of the tracker together with other methods. More details can be found in: https://github.com/LiYunfengLYF/KF_in_underwater_trackers.

VQNet 2.0: A New Generation Machine Learning Framework that Unifies Classical and Quantum

With the rapid development of classical and quantum machine learning, a large number of machine learning frameworks have been proposed. However, existing machine learning frameworks usually only focus on classical or quantum, rather than both. Therefore, based on VQNet 1.0, we further propose VQNet 2.0, a new generation of unified classical and quantum machine learning framework that supports hybrid optimization. The core library of the framework is implemented in C++, and the user level is implemented in Python, and it supports deployment on quantum and classical hardware. In this article, we analyze the development trend of the new generation machine learning framework and introduce the design principles of VQNet 2.0 in detail: unity, practicality, efficiency, and compatibility, as well as full particulars of implementation. We illustrate the functions of VQNet 2.0 through several basic applications, including classical convolutional neural networks, quantum autoencoders, hybrid classical-quantum networks, etc. After that, through extensive experiments, we demonstrate that the operation speed of VQNet 2.0 is higher than the comparison method. Finally, through extensive experiments, we demonstrate that VQNet 2.0 can deploy on different hardware platforms, the overall calculation speed is faster than the comparison method. It also can be mixed and optimized with quantum circuits composed of multiple quantum computing libraries.

Investigation of Network Architecture for Multimodal Head-and-Neck Tumor Segmentation

Inspired by the recent success of Transformers for Natural Language Processing and vision Transformer for Computer Vision, many researchers in the medical imaging community have flocked to Transformer-based networks for various main stream medical tasks such as classification, segmentation, and estimation. In this study, we analyze, two recently published Transformer-based network architectures for the task of multimodal head-and-tumor segmentation and compare their performance to the de facto standard 3D segmentation network - the nnU-Net. Our results showed that modeling long-range dependencies may be helpful in cases where large structures are present and/or large field of view is needed. However, for small structures such as head-and-neck tumor, the convolution-based U-Net architecture seemed to perform well, especially when training dataset is small and computational resource is limited.