In recent years, sparse voxel-based methods have become the state-of-the-arts for 3D semantic segmentation of indoor scenes, thanks to the powerful 3D CNNs. Nevertheless, being oblivious to the underlying geometry, voxel-based methods suffer from ambiguous features on spatially close objects and struggle with handling complex and irregular geometries due to the lack of geodesic information. In view of this, we present Voxel-Mesh Network (VMNet), a novel 3D deep architecture that operates on the voxel and mesh representations leveraging both the Euclidean and geodesic information. Intuitively, the Euclidean information extracted from voxels can offer contextual cues representing interactions between nearby objects, while the geodesic information extracted from meshes can help separate objects that are spatially close but have disconnected surfaces. To incorporate such information from the two domains, we design an intra-domain attentive module for effective feature aggregation and an inter-domain attentive module for adaptive feature fusion. Experimental results validate the effectiveness of VMNet: specifically, on the challenging ScanNet dataset for large-scale segmentation of indoor scenes, it outperforms the state-of-the-art SparseConvNet and MinkowskiNet (74.6% vs 72.5% and 73.6% in mIoU) with a simpler network structure (17M vs 30M and 38M parameters). Code release: https://github.com/hzykent/VMNet
In medical image segmentation, it is difficult to mark ambiguous areas accurately with binary masks, especially when dealing with small lesions. Therefore, it is a challenge for radiologists to reach a consensus by using binary masks under the condition of multiple annotations. However, these areas may contain anatomical structures that are conducive to diagnosis. Uncertainty is introduced to study these situations. Nevertheless, the uncertainty is usually measured by the variances between predictions in a multiple trial way. It is not intuitive, and there is no exact correspondence in the image. Inspired by image matting, we introduce matting as a soft segmentation method and a new perspective to deal with and represent uncertain regions into medical scenes, namely medical matting. More specifically, because there is no available medical matting dataset, we first labeled two medical datasets with alpha matte. Secondly, the matting method applied to the natural image is not suitable for the medical scene, so we propose a new architecture to generate binary masks and alpha matte in a row. Thirdly, the uncertainty map is introduced to highlight the ambiguous regions from the binary results and improve the matting performance. Evaluated on these datasets, the proposed model outperformed state-of-the-art matting algorithms by a large margin, and alpha matte is proved to be a more efficient labeling form than a binary mask.
In this letter, we propose a multi-task over-theair federated learning (MOAFL) framework, where multiple learning tasks share edge devices for data collection and learning models under the coordination of a edge server (ES). Specially, the model updates for all the tasks are transmitted and superpositioned concurrently over a non-orthogonal uplink channel via over-the-air computation, and the aggregation results of all the tasks are reconstructed at the ES through an extended version of the turbo compressed sensing algorithm. Both the convergence analysis and numerical results demonstrate that the MOAFL framework can significantly reduce the uplink bandwidth consumption of multiple tasks without causing substantial learning performance degradation.
Optical wireless communication (OWC) meets the demands of the future six-generation mobile network (6G) as it operates at several hundreds of Terahertz and has the potential to enable data rate in the order of Tbps. However, most beam-steering OWC technologies require high-accuracy positioning and high-speed control. Resonant beam communication (RBCom), as one kind of non-positioning OWC technologies, has been proposed for high-rate mobile communications. The mobility of RBCom relies on its self-alignment characteristic where no positioning is required. In a previous study, an external-cavity second-harmonic-generation (SHG) RBCom system has been proposed for eliminating the echo interference inside the resonator. However, its energy conversion efficiency and complexity are of concern. In this paper, we propose an intra-cavity SHG RBCom system to simplify the system design and improve the energy conversion efficiency. We elaborate the system structure and establish an analytical model. Numerical results show that the energy consumption of the proposed intra-cavity design is reduced to reach the same level of channel capacity at the receiver compared with the external-cavity one.
Whether it be for results summarization, or the analysis of classifier fusion, some means to compare different classifiers can often provide illuminating insight into their behaviour, (dis)similarity or complementarity. We propose a simple method to derive 2D representation from detection scores produced by an arbitrary set of binary classifiers in response to a common dataset. Based upon rank correlations, our method facilitates a visual comparison of classifiers with arbitrary scores and with close relation to receiver operating characteristic (ROC) and detection error trade-off (DET) analyses. While the approach is fully versatile and can be applied to any detection task, we demonstrate the method using scores produced by automatic speaker verification and voice anti-spoofing systems. The former are produced by a Gaussian mixture model system trained with VoxCeleb data whereas the latter stem from submissions to the ASVspoof 2019 challenge.
Unsupervised pretraining has recently proven beneficial for computer vision tasks, including object detection. However, previous self-supervised approaches are not designed to handle a key aspect of detection: localizing objects. Here, we present DETReg, an unsupervised pretraining approach for object DEtection with TRansformers using Region priors. Motivated by the two tasks underlying object detection: localization and categorization, we combine two complementary signals for self-supervision. For an object localization signal, we use pseudo ground truth object bounding boxes from an off-the-shelf unsupervised region proposal method, Selective Search, which does not require training data and can detect objects at a high recall rate and very low precision. The categorization signal comes from an object embedding loss that encourages invariant object representations, from which the object category can be inferred. We show how to combine these two signals to train the Deformable DETR detection architecture from large amounts of unlabeled data. DETReg improves the performance over competitive baselines and previous self-supervised methods on standard benchmarks like MS COCO and PASCAL VOC. DETReg also outperforms previous supervised and unsupervised baseline approaches on low-data regime when trained with only 1%, 2%, 5%, and 10% of the labeled data on MS COCO. For code and pretrained models, visit the project page at https://amirbar.net/detreg
In forming learning objectives, one oftentimes needs to aggregate a set of individual values to a single output. Such cases occur in the aggregate loss, which combines individual losses of a learning model over each training sample, and in the individual loss for multi-label learning, which combines prediction scores over all class labels. In this work, we introduce the sum of ranked range (SoRR) as a general approach to form learning objectives. A ranked range is a consecutive sequence of sorted values of a set of real numbers. The minimization of SoRR is solved with the difference of convex algorithm (DCA). We explore two applications in machine learning of the minimization of the SoRR framework, namely the AoRR aggregate loss for binary/multi-class classification at the sample level and the TKML individual loss for multi-label/multi-class classification at the label level. A combination loss of AoRR and TKML is proposed as a new learning objective for improving the robustness of multi-label learning in the face of outliers in sample and labels alike. Our empirical results highlight the effectiveness of the proposed optimization frameworks and demonstrate the applicability of proposed losses using synthetic and real data sets.
When a human asks questions online, or when a conversational virtual agent asks human questions, questions triggering emotions or with details might more likely to get responses or answers. we explore how to automatically rewrite natural language questions to improve the response rate from people. In particular, a new task of Visual Question Rewriting(VQR) task is introduced to explore how visual information can be used to improve the new questions. A data set containing around 4K bland questions, attractive questions and images triples is collected. We developed some baseline sequence to sequence models and more advanced transformer based models, which take a bland question and a related image as input and output a rewritten question that is expected to be more attractive. Offline experiments and mechanical Turk based evaluations show that it is possible to rewrite bland questions in a more detailed and attractive way to increase the response rate, and images can be helpful.
Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbation can completely change prediction result. The vulnerability has led to a surge of research in this direction, including adversarial attacks on object detection networks. However, previous studies are dedicated to attacking anchor-based object detectors. In this paper, we present the first adversarial attack on anchor-free object detectors. It conducts category-wise, instead of previously instance-wise, attacks on object detectors, and leverages high-level semantic information to efficiently generate transferable adversarial examples, which can also be transferred to attack other object detectors, even anchor-based detectors such as Faster R-CNN. Experimental results on two benchmark datasets demonstrate that our proposed method achieves state-of-the-art performance and transferability.