Few researches have studied simultaneous detection of smoke and flame accompanying fires due to their different physical natures that lead to uncertain fluid patterns. In this study, we collect a large image data set to re-label them as a multi-label image classification problem so as to identify smoke and flame simultaneously. In order to solve the generalization ability of the detection model on account of the movable fluid objects with uncertain shapes like fire and smoke, and their not compactible natures as well as the complex backgrounds with high variations, we propose a data augment method by random image stitch to deploy resizing, deforming, position variation, and background altering so as to enlarge the view of the learner. Moreover, we propose a self-learning data augment method by using the class activation map to extract the highly trustable region as new data source of positive examples to further enhance the data augment. By the mutual reinforcement between the data augment and the detection model that are performed iteratively, both modules make progress in an evolutionary manner. Experiments show that the proposed method can effectively improve the generalization performance of the model for concurrent smoke and fire detection.
相關內容
Establishing dense correspondences across semantically similar images is one of the challenging tasks due to the significant intra-class variations and background clutters. To solve these problems, numerous methods have been proposed, focused on learning feature extractor or cost aggregation independently, which yields sub-optimal performance. In this paper, we propose a novel framework for jointly learning feature extraction and cost aggregation for semantic correspondence. By exploiting the pseudo labels from each module, the networks consisting of feature extraction and cost aggregation modules are simultaneously learned in a boosting fashion. Moreover, to ignore unreliable pseudo labels, we present a confidence-aware contrastive loss function for learning the networks in a weakly-supervised manner. We demonstrate our competitive results on standard benchmarks for semantic correspondence.
In this paper, we propose M$^2$BEV, a unified framework that jointly performs 3D object detection and map segmentation in the Birds Eye View~(BEV) space with multi-camera image inputs. Unlike the majority of previous works which separately process detection and segmentation, M$^2$BEV infers both tasks with a unified model and improves efficiency. M$^2$BEV efficiently transforms multi-view 2D image features into the 3D BEV feature in ego-car coordinates. Such BEV representation is important as it enables different tasks to share a single encoder. Our framework further contains four important designs that benefit both accuracy and efficiency: (1) An efficient BEV encoder design that reduces the spatial dimension of a voxel feature map. (2) A dynamic box assignment strategy that uses learning-to-match to assign ground-truth 3D boxes with anchors. (3) A BEV centerness re-weighting that reinforces with larger weights for more distant predictions, and (4) Large-scale 2D detection pre-training and auxiliary supervision. We show that these designs significantly benefit the ill-posed camera-based 3D perception tasks where depth information is missing. M$^2$BEV is memory efficient, allowing significantly higher resolution images as input, with faster inference speed. Experiments on nuScenes show that M$^2$BEV achieves state-of-the-art results in both 3D object detection and BEV segmentation, with the best single model achieving 42.5 mAP and 57.0 mIoU in these two tasks, respectively.
The latest biological findings discover that the motionless 'lock-and-key' theory is no longer applicable and the flexibility of both the receptor and ligand plays a significant role in helping understand the principles of the binding affinity prediction. Based on this mechanism, molecular dynamics (MD) simulations have been invented as a useful tool to investigate the dynamical properties of this molecular system. However, the computational expenditure prohibits the growth of reported protein trajectories. To address this insufficiency, we present a novel spatial-temporal pre-training protocol, PretrainMD, to grant the protein encoder the capacity to capture the time-dependent geometric mobility along MD trajectories. Specifically, we introduce two sorts of self-supervised learning tasks: an atom-level denoising generative task and a protein-level snapshot ordering task. We validate the effectiveness of PretrainMD through the PDBbind dataset for both linear-probing and fine-tuning. Extensive experiments show that our PretrainMD exceeds most state-of-the-art methods and achieves comparable performance. More importantly, through visualization we discover that the learned representations by pre-training on MD trajectories without any label from the downstream task follow similar patterns of the magnitude of binding affinities. This strongly aligns with the fact that the motion of the interactions of protein and ligand maintains the key information of their binding. Our work provides a promising perspective of self-supervised pre-training for protein representations with very fine temporal resolutions and hopes to shed light on the further usage of MD simulations for the biomedical deep learning community.
Remote-sensing (RS) Change Detection (CD) aims to detect "changes of interest" from co-registered bi-temporal images. The performance of existing deep supervised CD methods is attributed to the large amounts of annotated data used to train the networks. However, annotating large amounts of remote sensing images is labor-intensive and expensive, particularly with bi-temporal images, as it requires pixel-wise comparisons by a human expert. On the other hand, we often have access to unlimited unlabeled multi-temporal RS imagery thanks to ever-increasing earth observation programs. In this paper, we propose a simple yet effective way to leverage the information from unlabeled bi-temporal images to improve the performance of CD approaches. More specifically, we propose a semi-supervised CD model in which we formulate an unsupervised CD loss in addition to the supervised Cross-Entropy (CE) loss by constraining the output change probability map of a given unlabeled bi-temporal image pair to be consistent under the small random perturbations applied on the deep feature difference map that is obtained by subtracting their latent feature representations. Experiments conducted on two publicly available CD datasets show that the proposed semi-supervised CD method can reach closer to the performance of supervised CD even with access to as little as 10% of the annotated training data. Code available at //github.com/wgcban/SemiCD
With the rapid growth of software, using third-party libraries (TPLs) has become increasingly popular. The prosperity of the library usage has provided the software engineers with handful of methods to facilitate and boost the program development. Unfortunately, it also poses great challenges as it becomes much more difficult to manage the large volume of libraries. Researches and studies have been proposed to detect and understand the TPLs in the software. However, most existing approaches rely on syntactic features, which are not robust when these features are changed or deliberately hidden by the adversarial parties. Moreover, these approaches typically model each of the imported libraries as a whole, therefore, cannot be applied to scenarios where the host software only partially uses the library code segments. To detect both fully and partially imported TPLs at the semantic level, we propose ModX, a framework that leverages novel program modularization techniques to decompose the program into finegrained functionality-based modules. By extracting both syntactic and semantic features, it measures the distance between modules to detect similar library module reuse in the program. Experimental results show that ModX outperforms other modularization tools by distinguishing more coherent program modules with 353% higher module quality scores and beats other TPL detection tools with on average 17% better in precision and 8% better in recall.
It's hard for neural MWP solvers to deal with tiny local variances. In MWP task, some local changes conserve the original semantic while the others may totally change the underlying logic. Currently, existing datasets for MWP task contain limited samples which are key for neural models to learn to disambiguate different kinds of local variances in questions and solve the questions correctly. In this paper, we propose a set of novel data augmentation approaches to supplement existing datasets with such data that are augmented with different kinds of local variances, and help to improve the generalization ability of current neural models. New samples are generated by knowledge guided entity replacement, and logic guided problem reorganization. The augmentation approaches are ensured to keep the consistency between the new data and their labels. Experimental results have shown the necessity and the effectiveness of our methods.
Cross-slide image analysis provides additional information by analysing the expression of different biomarkers as compared to a single slide analysis. These biomarker stained slides are analysed side by side, revealing unknown relations between them. During the slide preparation, a tissue section may be placed at an arbitrary orientation as compared to other sections of the same tissue block. The problem is compounded by the fact that tissue contents are likely to change from one section to the next and there may be unique artefacts on some of the slides. This makes registration of each section to a reference section of the same tissue block an important pre-requisite task before any cross-slide analysis. We propose a deep feature based registration (DFBR) method which utilises data-driven features to estimate the rigid transformation. We adopted a multi-stage strategy for improving the quality of registration. We also developed a visualisation tool to view registered pairs of WSIs at different magnifications. With the help of this tool, one can apply a transformation on the fly without the need to generate transformed source WSI in a pyramidal form. We compared the performance of data-driven features with that of hand-crafted features on the COMET dataset. Our approach can align the images with low registration errors. Generally, the success of non-rigid registration is dependent on the quality of rigid registration. To evaluate the efficacy of the DFBR method, the first two steps of the ANHIR winner's framework are replaced with our DFBR to register challenge provided image pairs. The modified framework produces comparable results to that of challenge winning team.
Most object recognition approaches predominantly focus on learning discriminative visual patterns while overlooking the holistic object structure. Though important, structure modeling usually requires significant manual annotations and therefore is labor-intensive. In this paper, we propose to "look into object" (explicitly yet intrinsically model the object structure) through incorporating self-supervisions into the traditional framework. We show the recognition backbone can be substantially enhanced for more robust representation learning, without any cost of extra annotation and inference speed. Specifically, we first propose an object-extent learning module for localizing the object according to the visual patterns shared among the instances in the same category. We then design a spatial context learning module for modeling the internal structures of the object, through predicting the relative positions within the extent. These two modules can be easily plugged into any backbone networks during training and detached at inference time. Extensive experiments show that our look-into-object approach (LIO) achieves large performance gain on a number of benchmarks, including generic object recognition (ImageNet) and fine-grained object recognition tasks (CUB, Cars, Aircraft). We also show that this learning paradigm is highly generalizable to other tasks such as object detection and segmentation (MS COCO). Project page: //github.com/JDAI-CV/LIO.
Modern neural network training relies heavily on data augmentation for improved generalization. After the initial success of label-preserving augmentations, there has been a recent surge of interest in label-perturbing approaches, which combine features and labels across training samples to smooth the learned decision surface. In this paper, we propose a new augmentation method that leverages the first and second moments extracted and re-injected by feature normalization. We replace the moments of the learned features of one training image by those of another, and also interpolate the target labels. As our approach is fast, operates entirely in feature space, and mixes different signals than prior methods, one can effectively combine it with existing augmentation methods. We demonstrate its efficacy across benchmark data sets in computer vision, speech, and natural language processing, where it consistently improves the generalization performance of highly competitive baseline networks.
Inspired by recent development of artificial satellite, remote sensing images have attracted extensive attention. Recently, noticeable progress has been made in scene classification and target detection.However, it is still not clear how to describe the remote sensing image content with accurate and concise sentences. In this paper, we investigate to describe the remote sensing images with accurate and flexible sentences. First, some annotated instructions are presented to better describe the remote sensing images considering the special characteristics of remote sensing images. Second, in order to exhaustively exploit the contents of remote sensing images, a large-scale aerial image data set is constructed for remote sensing image caption. Finally, a comprehensive review is presented on the proposed data set to fully advance the task of remote sensing caption. Extensive experiments on the proposed data set demonstrate that the content of the remote sensing image can be completely described by generating language descriptions. The data set is available at //github.com/2051/RSICD_optimal
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191