Abstract:The existing change detection(CD) methods can be summarized as the visual-first change detection (ViFi-CD) paradigm, which first extracts change features from visual differences and then assigns them specific semantic information. However, CD is essentially dependent on change regions of interest (CRoIs), meaning that the CD results are directly determined by the semantics changes of interest, making its primary image factor semantic of interest rather than visual. The ViFi-CD paradigm can only assign specific semantics of interest to specific change features extracted from visual differences, leading to the inevitable omission of potential CRoIs and the inability to adapt to different CRoI CD tasks. In other words, changes in other CRoIs cannot be detected by the ViFi-CD method without retraining the model or significantly modifying the method. This paper introduces a new CD paradigm, the semantic-first CD (SeFi-CD) paradigm. The core idea of SeFi-CD is to first perceive the dynamic semantics of interest and then visually search for change features related to the semantics. Based on the SeFi-CD paradigm, we designed Anything You Want Change Detection (AUWCD). Experiments on public datasets demonstrate that the AUWCD outperforms the current state-of-the-art CD methods, achieving an average F1 score 5.01\% higher than that of these advanced supervised baselines on the SECOND dataset, with a maximum increase of 13.17\%. The proposed SeFi-CD offers a novel CD perspective and approach.
Abstract:The remote sensing image intelligence understanding model is undergoing a new profound paradigm shift which has been promoted by multi-modal large language model (MLLM), i.e. from the paradigm learning a domain model (LaDM) shifts to paradigm learning a pre-trained general foundation model followed by an adaptive domain model (LaGD). Under the new LaGD paradigm, the old datasets, which have led to advances in RSI intelligence understanding in the last decade, are no longer suitable for fire-new tasks. We argued that a new dataset must be designed to lighten tasks with the following features: 1) Generalization: training model to learn shared knowledge among tasks and to adapt to different tasks; 2) Understanding complex scenes: training model to understand the fine-grained attribute of the objects of interest, and to be able to describe the scene with natural language; 3) Reasoning: training model to be able to realize high-level visual reasoning. In this paper, we designed a high-quality, diversified, and unified multimodal instruction-following dataset for RSI understanding produced by GPT-4V and existing datasets, which we called RS-GPT4V. To achieve generalization, we used a (Question, Answer) which was deduced from GPT-4V via instruction-following to unify the tasks such as captioning and localization; To achieve complex scene, we proposed a hierarchical instruction description with local strategy in which the fine-grained attributes of the objects and their spatial relationships are described and global strategy in which all the local information are integrated to yield detailed instruction descript; To achieve reasoning, we designed multiple-turn QA pair to provide the reasoning ability for a model. The empirical results show that the fine-tuned MLLMs by RS-GPT4V can describe fine-grained information. The dataset is available at: https://github.com/GeoX-Lab/RS-GPT4V.
Abstract:Scene graph generation (SGG) in satellite imagery (SAI) benefits promoting intelligent understanding of geospatial scenarios from perception to cognition. In SAI, objects exhibit great variations in scales and aspect ratios, and there exist rich relationships between objects (even between spatially disjoint objects), which makes it necessary to holistically conduct SGG in large-size very-high-resolution (VHR) SAI. However, the lack of SGG datasets with large-size VHR SAI has constrained the advancement of SGG in SAI. Due to the complexity of large-size VHR SAI, mining triplets <subject, relationship, object> in large-size VHR SAI heavily relies on long-range contextual reasoning. Consequently, SGG models designed for small-size natural imagery are not directly applicable to large-size VHR SAI. To address the scarcity of datasets, this paper constructs a large-scale dataset for SGG in large-size VHR SAI with image sizes ranging from 512 x 768 to 27,860 x 31,096 pixels, named RSG, encompassing over 210,000 objects and more than 400,000 triplets. To realize SGG in large-size VHR SAI, we propose a context-aware cascade cognition (CAC) framework to understand SAI at three levels: object detection (OBD), pair pruning and relationship prediction. As a fundamental prerequisite for SGG in large-size SAI, a holistic multi-class object detection network (HOD-Net) that can flexibly integrate multi-scale contexts is proposed. With the consideration that there exist a huge amount of object pairs in large-size SAI but only a minority of object pairs contain meaningful relationships, we design a pair proposal generation (PPG) network via adversarial reconstruction to select high-value pairs. Furthermore, a relationship prediction network with context-aware messaging (RPCM) is proposed to predict the relationship types of these pairs.
Abstract:The tokenizer, as one of the fundamental components of large models, has long been overlooked or even misunderstood in visual tasks. One key factor of the great comprehension power of the large language model is that natural language tokenizers utilize meaningful words or subwords as the basic elements of language. In contrast, mainstream visual tokenizers, represented by patch-based methods such as Patch Embed, rely on meaningless rectangular patches as basic elements of vision, which cannot serve as effectively as words or subwords in language. Starting from the essence of the tokenizer, we defined semantically independent regions (SIRs) for vision. We designed a simple HOmogeneous visual tOKenizer: HOOK. HOOK mainly consists of two modules: the Object Perception Module (OPM) and the Object Vectorization Module (OVM). To achieve homogeneity, the OPM splits the image into 4*4 pixel seeds and then utilizes the attention mechanism to perceive SIRs. The OVM employs cross-attention to merge seeds within the same SIR. To achieve adaptability, the OVM defines a variable number of learnable vectors as cross-attention queries, allowing for the adjustment of token quantity. We conducted experiments on the NWPU-RESISC45, WHU-RS19 classification dataset, and GID5 segmentation dataset for sparse and dense tasks. The results demonstrate that the visual tokens obtained by HOOK correspond to individual objects, which demonstrates homogeneity. HOOK outperformed Patch Embed by 6\% and 10\% in the two tasks and achieved state-of-the-art performance compared to the baselines used for comparison. Compared to Patch Embed, which requires more than one hundred tokens for one image, HOOK requires only 6 and 8 tokens for sparse and dense tasks, respectively, resulting in efficiency improvements of 1.5 to 2.8 times. The code is available at https://github.com/GeoX-Lab/Hook.
Abstract:Accurate traffic forecasting is a fundamental problem in intelligent transportation systems and learning long-range traffic representations with key information through spatiotemporal graph neural networks (STGNNs) is a basic assumption of current traffic flow prediction models. However, due to structural limitations, existing STGNNs can only utilize short-range traffic flow data; therefore, the models cannot adequately learn the complex trends and periodic features in traffic flow. Besides, it is challenging to extract the key temporal information from the long historical traffic series and obtain a compact representation. To solve the above problems, we propose a novel LSTTN (Long-Short Term Transformer-based Network) framework comprehensively considering the long- and short-term features in historical traffic flow. First, we employ a masked subseries Transformer to infer the content of masked subseries from a small portion of unmasked subseries and their temporal context in a pretraining manner, forcing the model to efficiently learn compressed and contextual subseries temporal representations from long historical series. Then, based on the learned representations, long-term trend is extracted by using stacked 1D dilated convolution layers, and periodic features are extracted by dynamic graph convolution layers. For the difficulties in making time-step level prediction, LSTTN adopts a short-term trend extractor to learn fine-grained short-term temporal features. Finally, LSTTN fuses the long-term trend, periodic features and short-term features to obtain the prediction results. Experiments on four real-world datasets show that in 60-minute-ahead long-term forecasting, the LSTTN model achieves a minimum improvement of 5.63\% and a maximum improvement of 16.78\% over baseline models. The source code is available at https://github.com/GeoX-Lab/LSTTN.
Abstract:Deep neural network-based Synthetic Aperture Radar (SAR) target recognition models are susceptible to adversarial examples. Current adversarial example generation methods for SAR imagery primarily operate in the 2D digital domain, known as image adversarial examples. Recent work, while considering SAR imaging scatter mechanisms, fails to account for the actual imaging process, rendering attacks in the three-dimensional physical domain infeasible, termed pseudo physics adversarial examples. To address these challenges, this paper proposes SAR-AE-SFP-Attack, a method to generate real physics adversarial examples by altering the scattering feature parameters of target objects. Specifically, we iteratively optimize the coherent energy accumulation of the target echo by perturbing the reflection coefficient and scattering coefficient in the scattering feature parameters of the three-dimensional target object, and obtain the adversarial example after echo signal processing and imaging processing in the RaySAR simulator. Experimental results show that compared to digital adversarial attack methods, SAR-AE-SFP Attack significantly improves attack efficiency on CNN-based models (over 30\%) and Transformer-based models (over 13\%), demonstrating significant transferability of attack effects across different models and perspectives.
Abstract:In the later training stages, further improvement of the models ability to determine changes relies on how well the change detection (CD) model learns hard cases; however, there are two additional challenges to learning hard case samples: (1) change labels are limited and tend to pointer only to foreground targets, yet hard case samples are prevalent in the background, which leads to optimizing the loss function focusing on the foreground targets and ignoring the background hard cases, which we call imbalance. (2) Complex situations, such as light shadows, target occlusion, and seasonal changes, induce hard case samples, and in the absence of both supervisory and scene information, it is difficult for the model to learn hard case samples directly to accurately obtain the feature representations of the change information, which we call missingness. We propose a Siamese foreground association-driven hard case sample optimization network (HSONet). To deal with this imbalance, we propose an equilibrium optimization loss function to regulate the optimization focus of the foreground and background, determine the hard case samples through the distribution of the loss values, and introduce dynamic weights in the loss term to gradually shift the optimization focus of the loss from the foreground to the background hard cases as the training progresses. To address this missingness, we understand hard case samples with the help of the scene context, propose the scene-foreground association module, use potential remote sensing spatial scene information to model the association between the target of interest in the foreground and the related context to obtain scene embedding, and apply this information to the feature reinforcement of hard cases. Experiments on four public datasets show that HSONet outperforms current state-of-the-art CD methods, particularly in detecting hard case samples.
Abstract:For a long time, due to the high heterogeneity in structure and semantics among various spatiotemporal modal data, the joint interpretation of multimodal spatiotemporal data has been an extremely challenging problem. The primary challenge resides in striking a trade-off between the cohesion and autonomy of diverse modalities, and this trade-off exhibits a progressively nonlinear nature as the number of modalities expands. We introduce the Language as Reference Framework (LaRF), a fundamental principle for constructing a multimodal unified model, aiming to strike a trade-off between the cohesion and autonomy among different modalities. We propose a multimodal spatiotemporal general artificial intelligence model, called AllSpark. Our model integrates thirteen different modalities into a unified framework, including 1D (text, code), 2D (RGB, infrared, SAR, multispectral, hyperspectral, tables, graphs, trajectory, oblique photography), and 3D (point clouds, videos) modalities. To achieve modal cohesion, AllSpark uniformly maps diverse modal features to the language modality. In addition, we design modality-specific prompts to guide multi-modal large language models in accurately perceiving multimodal data. To maintain modality autonomy, AllSpark introduces modality-specific encoders to extract the tokens of various spatiotemporal modalities. And modal bridge is employed to achieve dimensional projection from each modality to the language modality. Finally, observing a gap between the model's interpretation and downstream tasks, we designed task heads to enhance the model's generalization capability on specific downstream tasks. Experiments indicate that AllSpark achieves competitive accuracy in modalities such as RGB and trajectory compared to state-of-the-art models.
Abstract:Ranker and retriever are two important components in dense passage retrieval. The retriever typically adopts a dual-encoder model, where queries and documents are separately input into two pre-trained models, and the vectors generated by the models are used for similarity calculation. The ranker often uses a cross-encoder model, where the concatenated query-document pairs are input into a pre-trained model to obtain word similarities. However, the dual-encoder model lacks interaction between queries and documents due to its independent encoding, while the cross-encoder model requires substantial computational cost for attention calculation, making it difficult to obtain real-time retrieval results. In this paper, we propose a dense retrieval model called MD2PR based on multi-level distillation. In this model, we distill the knowledge learned from the cross-encoder to the dual-encoder at both the sentence level and word level. Sentence-level distillation enhances the dual-encoder on capturing the themes and emotions of sentences. Word-level distillation improves the dual-encoder in analysis of word semantics and relationships. As a result, the dual-encoder can be used independently for subsequent encoding and retrieval, avoiding the significant computational cost associated with the participation of the cross-encoder. Furthermore, we propose a simple dynamic filtering method, which updates the threshold during multiple training iterations to ensure the effective identification of false negatives and thus obtains a more comprehensive semantic representation space. The experimental results over two standard datasets show our MD2PR outperforms 11 baseline models in terms of MRR and Recall metrics.
Abstract:We apply the U-Net model for compressive light field synthesis. Compared to methods based on stacked CNN and iterative algorithms, this method offers better image quality, uniformity and less computation.