In 3D medical image segmentation, small targets segmentation is crucial for diagnosis but still faces challenges. In this paper, we propose the Axis Projection Attention UNet, named APAUNet, for 3D medical image segmentation, especially for small targets. Considering the large proportion of the background in the 3D feature space, we introduce a projection strategy to project the 3D features into three orthogonal 2D planes to capture the contextual attention from different views. In this way, we can filter out the redundant feature information and mitigate the loss of critical information for small lesions in 3D scans. Then we utilize a dimension hybridization strategy to fuse the 3D features with attention from different axes and merge them by a weighted summation to adaptively learn the importance of different perspectives. Finally, in the APA Decoder, we concatenate both high and low resolution features in the 2D projection process, thereby obtaining more precise multi-scale information, which is vital for small lesion segmentation. Quantitative and qualitative experimental results on two public datasets (BTCV and MSD) demonstrate that our proposed APAUNet outperforms the other methods. Concretely, our APAUNet achieves an average dice score of 87.84 on BTCV, 84.48 on MSD-Liver and 69.13 on MSD-Pancreas, and significantly surpass the previous SOTA methods on small targets.
This paper considers improving wireless communication and computation efficiency in federated learning (FL) via model quantization. In the proposed bitwidth FL scheme, edge devices train and transmit quantized versions of their local FL model parameters to a coordinating server, which, in turn, aggregates them into a quantized global model and synchronizes the devices. The goal is to jointly determine the bitwidths employed for local FL model quantization and the set of devices participating in FL training at each iteration. This problem is posed as an optimization problem whose goal is to minimize the training loss of quantized FL under a per-iteration device sampling budget and delay requirement. To derive the solution, an analytical characterization is performed in order to show how the limited wireless resources and induced quantization errors affect the performance of the proposed FL method. The analytical results show that the improvement of FL training loss between two consecutive iterations depends on the device selection and quantization scheme as well as on several parameters inherent to the model being learned. Given linear regression-based estimates of these model properties, it is shown that the FL training process can be described as a Markov decision process (MDP), and, then, a model-based reinforcement learning (RL) method is proposed to optimize action selection over iterations. Compared to model-free RL, this model-based RL approach leverages the derived mathematical characterization of the FL training process to discover an effective device selection and quantization scheme without imposing additional device communication overhead. Simulation results show that the proposed FL algorithm can reduce 29% and 63% convergence time compared to a model free RL method and the standard FL method, respectively.
Estimating accurate lane lines in 3D space remains challenging due to their sparse and slim nature. In this work, we propose the M^2-3DLaneNet, a Multi-Modal framework for effective 3D lane detection. Aiming at integrating complementary information from multi-sensors, M^2-3DLaneNet first extracts multi-modal features with modal-specific backbones, then fuses them in a unified Bird's-Eye View (BEV) space. Specifically, our method consists of two core components. 1) To achieve accurate 2D-3D mapping, we propose the top-down BEV generation. Within it, a Line-Restricted Deform-Attention (LRDA) module is utilized to effectively enhance image features in a top-down manner, fully capturing the slenderness features of lanes. After that, it casts the 2D pyramidal features into 3D space using depth-aware lifting and generates BEV features through pillarization. 2) We further propose the bottom-up BEV fusion, which aggregates multi-modal features through multi-scale cascaded attention, integrating complementary information from camera and LiDAR sensors. Sufficient experiments demonstrate the effectiveness of M^2-3DLaneNet, which outperforms previous state-of-the-art methods by a large margin, i.e., 12.1% F1-score improvement on OpenLane dataset.
In this paper, a semantic communication framework is proposed for textual data transmission. In the studied model, a base station (BS) extracts the semantic information from textual data, and transmits it to each user. The semantic information is modeled by a knowledge graph (KG) that consists of a set of semantic triples. After receiving the semantic information, each user recovers the original text using a graph-to-text generation model. To measure the performance of the considered semantic communication framework, a metric of semantic similarity (MSS) that jointly captures the semantic accuracy and completeness of the recovered text is proposed. Due to wireless resource limitations, the BS may not be able to transmit the entire semantic information to each user and satisfy the transmission delay constraint. Hence, the BS must select an appropriate resource block for each user as well as determine and transmit part of the semantic information to the users. As such, we formulate an optimization problem whose goal is to maximize the total MSS by jointly optimizing the resource allocation policy and determining the partial semantic information to be transmitted. To solve this problem, a proximal-policy-optimization-based reinforcement learning (RL) algorithm integrated with an attention network is proposed. The proposed algorithm can evaluate the importance of each triple in the semantic information using an attention network and then, build a relationship between the importance distribution of the triples in the semantic information and the total MSS. Compared to traditional RL algorithms, the proposed algorithm can dynamically adjust its learning rate thus ensuring convergence to a locally optimal solution.
Personalized federated learning (FL) facilitates collaborations between multiple clients to learn personalized models without sharing private data. The mechanism mitigates the statistical heterogeneity commonly encountered in the system, i.e., non-IID data over different clients. Existing personalized algorithms generally assume all clients volunteer for personalization. However, potential participants might still be reluctant to personalize models since they might not work well. In this case, clients choose to use the global model instead. To avoid making unrealistic assumptions, we introduce the personalization rate, measured as the fraction of clients willing to train personalized models, into federated settings and propose DyPFL. This dynamically personalized FL technique incentivizes clients to participate in personalizing local models while allowing the adoption of the global model when it performs better. We show that the algorithmic pipeline in DyPFL guarantees good convergence performance, allowing it to outperform alternative personalized methods in a broad range of conditions, including variation in heterogeneity, number of clients, local epochs, and batch sizes.
In recent years, the exponential increase in the demand of wireless data transmission rises the urgency for accurate spectrum sensing approaches to improve spectrum efficiency. The unreliability of conventional spectrum sensing methods by using measurements from a single secondary user (SU) has motivated research on cooperative spectrum sensing (CSS). In this work, we propose a vertical federated learning (VFL) framework to exploit the distributed features across multiple SUs without compromising data privacy. However, the repetitive training process in VFL faces the issue of high communication latency. To accelerate the training process, we propose a truncated vertical federated learning (T-VFL) algorithm, where the training latency is highly reduced by integrating the standard VFL algorithm with a channel-aware user scheduling policy. The convergence performance of T-VFL is provided via mathematical analysis and justified by simulation results. Moreover, to guarantee the convergence performance of the T-VFL algorithm, we conclude three design rules on the neural architectures used under the VFL framework, whose effectiveness is proved through simulations.
Real-world text applications often involve composing a wide range of text control operations, such as editing the text w.r.t. an attribute, manipulating keywords and structure, and generating new text of desired properties. Prior work typically learns/finetunes a language model (LM) to perform individual or specific subsets of operations. Recent research has studied combining operations in a plug-and-play manner, often with costly search or optimization in the complex sequence space. This paper proposes a new efficient approach for composable text operations in the compact latent space of text. The low-dimensionality and differentiability of the text latent vector allow us to develop an efficient sampler based on ordinary differential equations (ODEs) given arbitrary plug-in operators (e.g., attribute classifiers). By connecting pretrained LMs (e.g., GPT2) to the latent space through efficient adaption, we then decode the sampled vectors into desired text sequences. The flexible approach permits diverse control operators (sentiment, tense, formality, keywords, etc.) acquired using any relevant data from different domains. Experiments show that composing those operators within our approach manages to generate or edit high-quality text, substantially improving over previous methods in terms of generation quality and efficiency.
We present a new framework to reconstruct holistic 3D indoor scenes including both room background and indoor objects from single-view images. Existing methods can only produce 3D shapes of indoor objects with limited geometry quality because of the heavy occlusion of indoor scenes. To solve this, we propose an instance-aligned implicit function (InstPIFu) for detailed object reconstruction. Combining with instance-aligned attention module, our method is empowered to decouple mixed local features toward the occluded instances. Additionally, unlike previous methods that simply represents the room background as a 3D bounding box, depth map or a set of planes, we recover the fine geometry of the background via implicit representation. Extensive experiments on the e SUN RGB-D, Pix3D, 3D-FUTURE, and 3D-FRONT datasets demonstrate that our method outperforms existing approaches in both background and foreground object reconstruction. Our code and model will be made publicly available.
Recent progress on 3D scene understanding has explored visual grounding (3DVG) to localize a target object through a language description. However, existing methods only consider the dependency between the entire sentence and the target object, thus ignoring fine-grained relationships between contexts and non-target ones. In this paper, we extend 3DVG to a more reliable and explainable task, called 3D Phrase Aware Grounding (3DPAG). The 3DPAG task aims to localize the target object in the 3D scenes by explicitly identifying all phrase-related objects and then conducting reasoning according to contextual phrases. To tackle this problem, we label about 400K phrase-level annotations from 170K sentences in available 3DVG datasets, i.e., Nr3D, Sr3D and ScanRefer. By tapping on these developed datasets, we propose a novel framework, i.e., PhraseRefer, which conducts phrase-aware and object-level representation learning through phrase-object alignment optimization as well as phrase-specific pre-training. In our setting, we extend previous 3DVG methods to the phrase-aware scenario and provide metrics to measure the explainability of the 3DPAG task. Extensive results confirm that 3DPAG effectively boosts the 3DVG, and PhraseRefer achieves state-of-the-arts across three datasets, i.e., 63.0%, 54.4% and 55.5% overall accuracy on Sr3D, Nr3D and ScanRefer, respectively.
This paper studies a new multi-device edge artificial-intelligent (AI) system, which jointly exploits the AI model split inference and integrated sensing and communication (ISAC) to enable low-latency intelligent services at the network edge. In this system, multiple ISAC devices perform radar sensing to obtain multi-view data, and then offload the quantized version of extracted features to a centralized edge server, which conducts model inference based on the cascaded feature vectors. Under this setup and by considering classification tasks, we measure the inference accuracy by adopting an approximate but tractable metric, namely discriminant gain, which is defined as the distance of two classes in the Euclidean feature space under normalized covariance. To maximize the discriminant gain, we first quantify the influence of the sensing, computation, and communication processes on it with a derived closed-form expression. Then, an end-to-end task-oriented resource management approach is developed by integrating the three processes into a joint design. This integrated sensing, computation, and communication (ISCC) design approach, however, leads to a challenging non-convex optimization problem, due to the complicated form of discriminant gain and the device heterogeneity in terms of channel gain, quantization level, and generated feature subsets. Remarkably, the considered non-convex problem can be optimally solved based on the sum-of-ratios method. This gives the optimal ISCC scheme, that jointly determines the transmit power and time allocation at multiple devices for sensing and communication, as well as their quantization bits allocation for computation distortion control. By using human motions recognition as a concrete AI inference task, extensive experiments are conducted to verify the performance of our derived optimal ISCC scheme.