With the increasing availability of depth sensors, multimodal frameworks that combine color information with depth data are attracting increasing interest. In the challenging task of semantic segmentation, depth maps allow to distinguish between similarly colored objects at different depths and provide useful geometric cues. On the other side, ground truth data for semantic segmentation is burdensome to be provided and thus domain adaptation is another significant research area. Specifically, we address the challenging source-free domain adaptation setting where the adaptation is performed without reusing source data. We propose MISFIT: MultImodal Source-Free Information fusion Transformer, a depth-aware framework which injects depth information into a segmentation module based on vision transformers at multiple stages, namely at the input, feature and output levels. Color and depth style transfer helps early-stage domain alignment while re-wiring self-attention between modalities creates mixed features allowing the extraction of better semantic content. Furthermore, a depth-based entropy minimization strategy is also proposed to adaptively weight regions at different distances. Our framework, which is also the first approach using vision transformers for source-free semantic segmentation, shows noticeable performance improvements with respect to standard strategies.
相關內容
Brain age estimation is clinically important as it can provide valuable information in the context of neurodegenerative diseases such as Alzheimer's. Population graphs, which include multimodal imaging information of the subjects along with the relationships among the population, have been used in literature along with Graph Convolutional Networks (GCNs) and have proved beneficial for a variety of medical imaging tasks. A population graph is usually static and constructed manually using non-imaging information. However, graph construction is not a trivial task and might significantly affect the performance of the GCN, which is inherently very sensitive to the graph structure. In this work, we propose a framework that learns a population graph structure optimized for the downstream task. An attention mechanism assigns weights to a set of imaging and non-imaging features (phenotypes), which are then used for edge extraction. The resulting graph is used to train the GCN. The entire pipeline can be trained end-to-end. Additionally, by visualizing the attention weights that were the most important for the graph construction, we increase the interpretability of the graph. We use the UK Biobank, which provides a large variety of neuroimaging and non-imaging phenotypes, to evaluate our method on brain age regression and classification. The proposed method outperforms competing static graph approaches and other state-of-the-art adaptive methods. We further show that the assigned attention scores indicate that there are both imaging and non-imaging phenotypes that are informative for brain age estimation and are in agreement with the relevant literature.
There is a strong incentive to develop computational pathology models to i) ease the burden of tissue typology annotation from whole slide histological images; ii) transfer knowledge, e.g., tissue class separability from the withheld source domain to the distributionally shifted unlabeled target domain, and simultaneously iii) detect Open Set samples, i.e., unseen novel categories not present in the training source domain. This paper proposes a highly practical setting by addressing the abovementioned challenges in one fell swoop, i.e., source-free Open Set domain adaptation (SF-OSDA), which addresses the situation where a model pre-trained on the inaccessible source dataset can be adapted on the unlabeled target dataset containing Open Set samples. The central tenet of our proposed method is distilling knowledge from a self-supervised vision transformer trained in the target domain. We propose a novel style-based data augmentation used as hard positives for self-training a vision transformer in the target domain, yielding strongly contextualized embedding. Subsequently, semantically similar target images are clustered while the source model provides their corresponding weak pseudo-labels with unreliable confidence. Furthermore, we propose cluster relative maximum logit score (CRMLS) to rectify the confidence of the weak pseudo-labels and compute weighted class prototypes in the contextualized embedding space that are utilized for adapting the source model on the target domain. Our method significantly outperforms the previous methods, including open set detection, test-time adaptation, and SF-OSDA methods, setting the new state-of-the-art on three public histopathological datasets of colorectal cancer (CRC) assessment- Kather-16, Kather-19, and CRCTP. Our code is available at //github.com/LTS5/Proto-SF-OSDA.
The ability to scene understanding in adverse visual conditions, e.g., nighttime, has sparked active research for RGB-Thermal (RGB-T) semantic segmentation. However, it is essentially hampered by two critical problems: 1) the day-night gap of RGB images is larger than that of thermal images, and 2) the class-wise performance of RGB images at night is not consistently higher or lower than that of thermal images. we propose the first test-time adaptation (TTA) framework, dubbed Night-TTA, to address the problems for nighttime RGBT semantic segmentation without access to the source (daytime) data during adaptation. Our method enjoys three key technical parts. Firstly, as one modality (e.g., RGB) suffers from a larger domain gap than that of the other (e.g., thermal), Imaging Heterogeneity Refinement (IHR) employs an interaction branch on the basis of RGB and thermal branches to prevent cross-modal discrepancy and performance degradation. Then, Class Aware Refinement (CAR) is introduced to obtain reliable ensemble logits based on pixel-level distribution aggregation of the three branches. In addition, we also design a specific learning scheme for our TTA framework, which enables the ensemble logits and three student logits to collaboratively learn to improve the quality of predictions during the testing phase of our Night TTA. Extensive experiments show that our method achieves state-of-the-art (SoTA) performance with a 13.07% boost in mIoU.
Existing methods of cross-modal domain adaptation for 3D semantic segmentation predict results only via 2D-3D complementarity that is obtained by cross-modal feature matching. However, as lacking supervision in the target domain, the complementarity is not always reliable. The results are not ideal when the domain gap is large. To solve the problem of lacking supervision, we introduce masked modeling into this task and propose a method Mx2M, which utilizes masked cross-modality modeling to reduce the large domain gap. Our Mx2M contains two components. One is the core solution, cross-modal removal and prediction (xMRP), which makes the Mx2M adapt to various scenarios and provides cross-modal self-supervision. The other is a new way of cross-modal feature matching, the dynamic cross-modal filter (DxMF) that ensures the whole method dynamically uses more suitable 2D-3D complementarity. Evaluation of the Mx2M on three DA scenarios, including Day/Night, USA/Singapore, and A2D2/SemanticKITTI, brings large improvements over previous methods on many metrics.
Leveraging the low-power, event-driven computation and the inherent temporal dynamics, spiking neural networks (SNNs) are potentially ideal solutions for processing dynamic and asynchronous signals from event-based sensors. However, due to the challenges in training and the restrictions in architectural design, there are limited examples of competitive SNNs in the realm of event-based dense prediction when compared to artificial neural networks (ANNs). In this paper, we present an efficient spiking encoder-decoder network designed for large-scale event-based semantic segmentation tasks. This is achieved by optimizing the encoder using a hierarchical search method. To enhance learning from dynamic event streams, we harness the inherent adaptive threshold of spiking neurons to modulate network activation. Moreover, we introduce a dual-path Spiking Spatially-Adaptive Modulation (SSAM) block, specifically designed to enhance the representation of sparse events, thereby considerably improving network performance. Our proposed network achieves a 72.57% mean intersection over union (MIoU) on the DDD17 dataset and a 57.22% MIoU on the recently introduced, larger DSEC-Semantic dataset. This performance surpasses the current state-of-the-art ANNs by 4%, whilst consuming significantly less computational resources. To the best of our knowledge, this is the first study demonstrating SNNs outperforming ANNs in demanding event-based semantic segmentation tasks, thereby establishing the vast potential of SNNs in the field of event-based vision. Our source code will be made publicly accessible.
The advent of high-resolution multispectral/hyperspectral sensors, LiDAR DSM (Digital Surface Model) information and many others has provided us with an unprecedented wealth of data for Earth Observation. Multimodal AI seeks to exploit those complementary data sources, particularly for complex tasks like semantic segmentation. While specialized architectures have been developed, they are highly complicated via significant effort in model design, and require considerable re-engineering whenever a new modality emerges. Recent trends in general-purpose multimodal networks have shown great potential to achieve state-of-the-art performance across multiple multimodal tasks with one unified architecture. In this work, we investigate the performance of PerceiverIO, one in the general-purpose multimodal family, in the remote sensing semantic segmentation domain. Our experiments reveal that this ostensibly universal network struggles with object scale variation in remote sensing images and fails to detect the presence of cars from a top-down view. To address these issues, even with extreme class imbalance issues, we propose a spatial and volumetric learning component. Specifically, we design a UNet-inspired module that employs 3D convolution to encode vital local information and learn cross-modal features simultaneously, while reducing network computational burden via the cross-attention mechanism of PerceiverIO. The effectiveness of the proposed component is validated through extensive experiments comparing it with other methods such as 2D convolution, and dual local module (\ie the combination of Conv2D 1x1 and Conv2D 3x3 inspired by UNetFormer). The proposed method achieves competitive results with specialized architectures like UNetFormer and SwinUNet, showing its potential to minimize network architecture engineering with a minimal compromise on the performance.
Sentence embeddings enable us to capture the semantic similarity of short texts. Most sentence embedding models are trained for general semantic textual similarity (STS) tasks. Therefore, to use sentence embeddings in a particular domain, the model must be adapted to it in order to achieve good results. Usually, this is done by fine-tuning the entire sentence embedding model for the domain of interest. While this approach yields state-of-the-art results, all of the model's weights are updated during fine-tuning, making this method resource-intensive. Therefore, instead of fine-tuning entire sentence embedding models for each target domain individually, we propose to train lightweight adapters. These domain-specific adapters do not require fine-tuning all underlying sentence embedding model parameters. Instead, we only train a small number of additional parameters while keeping the weights of the underlying sentence embedding model fixed. Training domain-specific adapters allows always using the same base model and only exchanging the domain-specific adapters to adapt sentence embeddings to a specific domain. We show that using adapters for parameter-efficient domain adaptation of sentence embeddings yields competitive performance within 1% of a domain-adapted, entirely fine-tuned sentence embedding model while only training approximately 3.6% of the parameters.
Video instance segmentation (VIS) is the task that requires simultaneously classifying, segmenting and tracking object instances of interest in video. Recent methods typically develop sophisticated pipelines to tackle this task. Here, we propose a new video instance segmentation framework built upon Transformers, termed VisTR, which views the VIS task as a direct end-to-end parallel sequence decoding/prediction problem. Given a video clip consisting of multiple image frames as input, VisTR outputs the sequence of masks for each instance in the video in order directly. At the core is a new, effective instance sequence matching and segmentation strategy, which supervises and segments instances at the sequence level as a whole. VisTR frames the instance segmentation and tracking in the same perspective of similarity learning, thus considerably simplifying the overall pipeline and is significantly different from existing approaches. Without bells and whistles, VisTR achieves the highest speed among all existing VIS models, and achieves the best result among methods using single model on the YouTube-VIS dataset. For the first time, we demonstrate a much simpler and faster video instance segmentation framework built upon Transformers, achieving competitive accuracy. We hope that VisTR can motivate future research for more video understanding tasks.
Recently, contrastive learning (CL) has emerged as a successful method for unsupervised graph representation learning. Most graph CL methods first perform stochastic augmentation on the input graph to obtain two graph views and maximize the agreement of representations in the two views. Despite the prosperous development of graph CL methods, the design of graph augmentation schemes -- a crucial component in CL -- remains rarely explored. We argue that the data augmentation schemes should preserve intrinsic structures and attributes of graphs, which will force the model to learn representations that are insensitive to perturbation on unimportant nodes and edges. However, most existing methods adopt uniform data augmentation schemes, like uniformly dropping edges and uniformly shuffling features, leading to suboptimal performance. In this paper, we propose a novel graph contrastive representation learning method with adaptive augmentation that incorporates various priors for topological and semantic aspects of the graph. Specifically, on the topology level, we design augmentation schemes based on node centrality measures to highlight important connective structures. On the node attribute level, we corrupt node features by adding more noise to unimportant node features, to enforce the model to recognize underlying semantic information. We perform extensive experiments of node classification on a variety of real-world datasets. Experimental results demonstrate that our proposed method consistently outperforms existing state-of-the-art baselines and even surpasses some supervised counterparts, which validates the effectiveness of the proposed contrastive framework with adaptive augmentation.
Deep Convolutional Neural Networks have pushed the state-of-the art for semantic segmentation provided that a large amount of images together with pixel-wise annotations is available. Data collection is expensive and a solution to alleviate it is to use transfer learning. This reduces the amount of annotated data required for the network training but it does not get rid of this heavy processing step. We propose a method of transfer learning without annotations on the target task for datasets with redundant content and distinct pixel distributions. Our method takes advantage of the approximate content alignment of the images between two datasets when the approximation error prevents the reuse of annotation from one dataset to another. Given the annotations for only one dataset, we train a first network in a supervised manner. This network autonomously learns to generate deep data representations relevant to the semantic segmentation. Then the images in the new dataset, we train a new network to generate a deep data representation that matches the one from the first network on the previous dataset. The training consists in a regression between feature maps and does not require any annotations on the new dataset. We show that this method reaches performances similar to a classic transfer learning on the PASCAL VOC dataset with synthetic transformations.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191