Composed image retrieval (CIR) aims to retrieve target images by combining a reference image with a modification text. Traditional CIR methods often struggle with feature-level multimodal fusion, leading to deviations from the original embedding space. To address this, we propose a Dual-Branch Multi-Scale Network (DMN) that integrates a combining branch and a complete text branch. To enhance the use of captions generated by advanced image captioning models for CIR, the DMN leverages an attribute-driven disentanglement layer to separate features into distinct latent factors and employs a dual-path multimodal fusion module for effective feature integration. Additionally, a multi-scale matching module incorporating both global and local matching strategies is introduced to enhance fine-grained feature discrimination. Experimental results on the FashionIQ, Shoes, and CIRR datasets demonstrate that our DMN model consistently outperforms state-of-the-art methods, achieving improvements of up to 1.43% in mean recall metrics.

Current emotional text-to-speech tasks have achieved high-quality emotional speech by incorporating emotion modules into text-to-speech models. However, there has been limited in-depth research on embedding emotion modules within TTS models, and the expression of emotion in synthesized speech is constrained by both the TTS model and the emotion module, often preventing optimal results. This paper presents a novel TTS model, Grad-DiT, based on the DiT architecture of diffusion models, aimed at enhancing the naturalness and expressiveness of TTS. Unlike traditional U-net architectures, Grad-DiT leverages the Transformer architecture to better capture contextual information in text, resulting in more natural speech generation. Building on this model, we propose Emo-DiT, which incorporates an Emotion Feature Reconstruction (EFR) module to enable the synthesis of speech with specific emotional expressions. Experimental results show that Grad-DiT not only surpasses existing TTS models in speech quality but also significantly improves real-time performance and inference speed. Compared to traditional emotion generation methods, Emo-DiT offers more precise emotional expression, enabling the synthesis of speech with distinctive emotional characteristics and providing substantial support for future applications in emotional speech synthesis.

Accurate and reliable sensor calibration is essential to fuse LiDAR and inertial measurements, which are usually available in robotic applications. In this article, we propose a novel LiDAR-IMU calibration method within the continuous-time batch-optimization framework, where the intrinsics of both sensors and the spatial-temporal extrinsics between sensors are calibrated without using calibration infrastructure, such as fiducial tags. Compared to discrete-time approaches, the continuous-time formulation has natural advantages for fusing high-rate measurements from LiDAR and IMU sensors. To improve efficiency and address degenerate motions, the following two observability-aware modules are leveraged: first, The information-theoretic data selection policy selectsonlythe most informative segments for calibration during data collection, which significantly improves the calibration efficiency by processing only the selected informative segments. Second, the observability-aware state update mechanism in nonlinear least-squares optimization updatesonlythe identifiable directions in the state space with truncated singular value decomposition, which enables accurate calibration results even under degenerate cases where informative data segments are not available. The proposed LiDAR-IMU calibration approach has been validated extensively in both simulated and real-world experiments with different robot platforms, demonstrating its high accuracy and repeatability in commonly-seen human-made environments.

In this paper, we propose a highly accurate continuous-time trajectory estimation framework dedicated to SLAM (Simultaneous Localization and Mapping) applications, which enables fuse high-frequency and asynchronous sensor data effectively. We apply the proposed framework in a 3D LiDAR-inertial system for evaluations. The proposed method adopts a non-rigid registration method for continuous-time trajectory estimation and simultaneously removing the motion distortion in LiDAR scans. Additionally, we propose a two-state continuous-time trajectory correction method to efficiently and efficiently tackle the computationally-intractable global optimization problem when loop closure happens. We examine the accuracy of the proposed approach on several publicly available datasets and the data we collected. The experimental results indicate that the proposed method outperforms the discrete-time methods regarding accuracy especially when aggressive motion occurs. Furthermore, we open source our code at https://github.com/APRIL-ZJU/clins to benefit research community.

Sensor calibration is the fundamental block for a multi-sensor fusion system. This paper presents an accurate and repeatable LiDAR-IMU calibration method (termed LI-Calib), to calibrate the 6-DOF extrinsic transformation between the 3D LiDAR and the Inertial Measurement Unit (IMU). Regarding the high data capture rate for LiDAR and IMU sensors, LI-Calib adopts a continuous-time trajectory formulation based on B-Spline, which is more suitable for fusing high-rate or asynchronous measurements than discrete-time based approaches. Additionally, LI-Calib decomposes the space into cells and identifies the planar segments for data association, which renders the calibration problem well-constrained in usual scenarios without any artificial targets. We validate the proposed calibration approach on both simulated and real-world experiments. The results demonstrate the high accuracy and good repeatability of the proposed method in common human-made scenarios. To benefit the research community, we open-source our code at https://github.com/APRIL-ZJU/lidar_IMU_calib.

Self-localization is an essential technology for autonomous vehicles. Building robust odometry in a GPS-denied environment is still challenging, especially when LiDAR and camera are uninformative. In this paper, We propose a learning-based approach to cure the drift of gyroscope for vehicle localization. For consumer-level MEMS gyroscope (stability ~10° /h), our GyroNet can estimate the error of each measurement. For high-precision Fiber optics Gyroscope (stability ~0.05° /h), we build a FoGNet which can obtain its drift by observing data in a long time window. We perform comparative experiments on publicly available datasets. The results demonstrate that our GyroNet can get higher precision angular velocity than traditional digital filters and static initialization methods. In the vehicle localization, the FoGNet can effectively correct the small drift of the Fiber optics Gyroscope (FoG) and can achieve better results than the state-of-the-art method.

As sensornets are increasingly being deployed in mission-critical applications, it becomes imperative that we consider application QoS requirements in in-network processing (INP). Toward understanding the complexity of joint QoS and INP optimization, we study the problem of jointly optimizing packet packing (i.e., aggregating shorter packets into longer ones) and the timeliness of data delivery. We identify the conditions under which the problem is strong NP-hard, and we find that the problem complexity heavily depends on aggregation constraints (in particular, maximum packet size and reaggregation tolerance) instead of network and traffic properties. For cases when the problem is NP-hard, we show that there is no polynomial-time approximation scheme (PTAS); for cases when the problem can be solved in polynomial time, we design polynomial time, offline algorithms for finding the optimal packet packing schemes. To understand the impact of joint QoS and INP optimization on sensornet performance, we design a distributed, online protocol tPack that schedules packet transmissions to maximize the local utility of packet packing at each node. Using a testbed of 130 TelosB motes, we experimentally evaluate the properties of tPack. We find that jointly optimizing data delivery timeliness and packet packing and considering real-world aggregation constraints significantly improve network performance. Our findings shed light on the challenges, benefits, and solutions of joint QoS and INP optimization, and they also suggest open problems for future research.

There are many security threats in WSN, such as malicious nodes on the transmission paths dropping, fabricating or tampering the forwarded messages. Most of the existing security methods relied on special hardware facilities, mechanism of node monitoring, encryption and authentication technology, which greatly increase the sensorpsilas price or computing and communicating cost in WSN. In this paper, we propose a new method, named as DESCM, which is based on source coding and multi-path transmission. Theoretical analysis shows that DESCM can detect and locate malicious nodes with high probability. Comparing with other methods, DESCM does not need special hardware, encryption or authentication technology, which can detect the malicious nodes with lower cost.

As sensornets are increasingly being deployed in mission-critical applications, it becomes imperative that we consider application QoS requirements in in-network processing (INP). Towards understanding the complexity of joint QoS and INP optimization, we study the problem of jointly optimizing packet packing (i.e., aggregating shorter packets into longer ones) and the timeliness of data delivery. We identify the conditions under which the problem is strong NP-hard, and we find that the problem complexity heavily depends on aggregation constraints (in particular, maximum packet size and re-aggregation tolerance) instead of network and traffic properties. For cases when the problem is NP-hard, we show that there is no polynomial-time approximation scheme (PTAS); for cases when the problem can be solved in polynomial time, we design polynomial time, offline algorithms for finding the optimal packet packing schemes. To understand the impact of joint QoS and INP optimization on sensornet performance, we design a distributed, online protocol \emph{tPack} that schedules packet transmissions to maximize the local utility of packet packing at each node. Using a testbed of 130 TelosB motes, we experimentally evaluate the properties of tPack. We find that jointly optimizing data delivery timeliness and packet packing significantly improve network performance. Our findings shed light on the challenges, benefits, and solutions of joint QoS and INP optimization, and they also suggest open problems for future research.

In group decision making, because the decision-makers usually represent different interest backgrounds, it is worth to study how to make the different decision makers coordinate and cooperate for aggregating group opinions. In this paper, based on the analytic hierarchy process, we propose a nonlinear program model to obtain consensus priority vector, and point that the model can make decision-makers reach consensus by improving compatibility of judgement matrices. Moreover, we use the genetic-simulated annealing algorithm to obtain its optimal solution. Finally, a numerical example is presented to illustrate the application of this method.

Similar to the concept of on-line auctions presented by Ron Lavi and Noam Nisan [3], this paper discusses pricing algorithms for on-line reverse auction which bidders arrive one by one and on-line buyer must be required to make a decision immediately about each bid as it is received. For online buyer in a reverse auction, we propose on-line mean pricing algorithm and on-line randomized pricing algorithm, and then prove that the two algorithms are competitive and incentive compatible. Moreover, as the bid prices concentrated in a small domain, by competitive analysis for the two algorithms, we find their merits which can avoid the results of purchasing failure or more cost caused by reservation price algorithm. Finally, an example is obtained to illustrate their application.