Search

Automated crop mapping through Satellite Image Time Series (SITS) has emerged as a crucial avenue for agricultural monitoring and management. However, due to the low resolution and unclear parcel boundaries, annotating pixel-level masks is exceptionally complex and time-consuming in SITS. This paper embraces the weakly supervised paradigm (i.e., only image-level categories available) to liberate the crop mapping task from the exhaustive annotation burden. The unique characteristics of SITS give rise to several challenges in weakly supervised learning: (1) noise perturbation from spatially neighboring regions, and (2) erroneous semantic bias from anomalous temporal periods. To address the above difficulties, we propose a novel method, termed exploring space-time perceptive clues (Exact). First, we introduce a set of spatial clues to explicitly capture the representative patterns of different crops from the most class-relative regions. Besides, we leverage the temporal-to-class interaction of the model to emphasize the contributions of pivotal clips, thereby enhancing the model perception for crop regions. Build upon the space-time perceptive clues, we derive the clue-based CAMs to effectively supervise the SITS segmentation network. Our method demonstrates impressive performance on various SITS benchmarks. Remarkably, the segmentation network trained on Exact-generated masks achieves 95% of its fully supervised performance, showing the bright promise of weakly supervised paradigm in crop mapping scenario. Our code will be publicly available., Comment: Under review. Code will be available at https://github.com/MiSsU-HH/Exact

This paper classifies near-miss traffic videos using the SlowFast deep neural network that mimics the characteristics of the slow and fast visual information processed by two different streams from the M (Magnocellular) and P (Parvocellular) cells of the human brain. The approach significantly improves the accuracy of the traffic near-miss video analysis and presents insights into human visual perception in traffic scenarios. Moreover, it contributes to traffic safety enhancements and provides novel perspectives on the potential cognitive errors in traffic accidents., Comment: Best Research Paper Award for Asia-Pacific Region, The 30th ITS World Congress 2024

Directly probing deep tissue activities from body surfaces offers a noninvasive approach to monitoring essential physiological processes1-3. However, this method is technically challenged by rapid signal attenuation toward the body surface and confounding motion artifacts4-6 primarily due to excessive contact impedance and mechanical mismatch with conventional electrodes. Herein, by formulating and directly spray coating biocompatible two-dimensional nanosheet ink onto the human body under ambient conditions, we create microscopically conformal and adaptive van der Waals thin films (VDWTFs) that seamlessly merge with non-Euclidean, hairy, and dynamically evolving body surfaces. Unlike traditional deposition methods, which often struggle with conformality and adaptability while retaining high electronic performance, this gentle process enables the formation of high-performance VDWTFs directly on the body surface under bio-friendly conditions, making it ideal for biological applications. This results in low-impedance electrically functionalized body surfaces (EFBS), enabling highly robust monitoring of biopotential and bioimpedance modulations associated with deep-tissue activities, such as blood circulation, muscle movements, and brain activities. Compared to commercial solutions, our VDWTF-EFBS exhibits nearly two-orders of magnitude lower contact impedance and substantially reduces the extrinsic motion artifacts, enabling reliable extraction of bioelectrical signals from irregular surfaces, such as unshaved human scalps. This advancement defines a technology for continuous, noninvasive monitoring of deep-tissue activities during routine body movements.

Retrieval-Augmented Generation (RAG) systems enhance large language models (LLMs) by integrating external knowledge, making them adaptable and cost-effective for various applications. However, the growing reliance on these systems also introduces potential security risks. In this work, we reveal a novel vulnerability, the retrieval prompt hijack attack (HijackRAG), which enables attackers to manipulate the retrieval mechanisms of RAG systems by injecting malicious texts into the knowledge database. When the RAG system encounters target questions, it generates the attacker's pre-determined answers instead of the correct ones, undermining the integrity and trustworthiness of the system. We formalize HijackRAG as an optimization problem and propose both black-box and white-box attack strategies tailored to different levels of the attacker's knowledge. Extensive experiments on multiple benchmark datasets show that HijackRAG consistently achieves high attack success rates, outperforming existing baseline attacks. Furthermore, we demonstrate that the attack is transferable across different retriever models, underscoring the widespread risk it poses to RAG systems. Lastly, our exploration of various defense mechanisms reveals that they are insufficient to counter HijackRAG, emphasizing the urgent need for more robust security measures to protect RAG systems in real-world deployments.

The JWST and ALMA have detected emission lines from the ionized interstellar medium (ISM), including [OII], [OIII], and hydrogen Balmer series lines, in some of the first galaxies at z>6. These measurements present an opportunity to better understand galaxy assembly histories and may allow important tests of state-of-the-art galaxy formation simulations. It is challenging, however, to model these lines in their proper cosmological context given the huge dynamic range in spatial scales involved. In order to meet this challenge, we introduce a novel sub-grid line emission modeling framework. The framework uses the high-z zoom-in simulation suite from the Feedback in Realistic Environments (FIRE) collaboration. The line emission signals from HII regions within each simulated FIRE galaxy are modeled using the semi-analytic HIILines code. A machine learning, Mixture Density Network, approach is then used to determine the conditional probability distribution for the line luminosity to stellar-mass ratio from the HII regions around each simulated stellar particle given its age, metallicity, and its galaxy's total stellar mass. This conditional probability distribution can then be applied to predict the line luminosities around stellar particles in lower resolution, yet larger volume cosmological simulations. As an example, we apply this approach to the Illustris-TNG simulations at z=6. The resulting predictions for the [OII], [OIII], and Balmer line luminosities as a function of star-formation rate agree well with current observations. Our predictions differ, however, from related work in the literature which lack detailed sub-grid ISM models. This highlights the importance of our multi-scale simulation modeling framework. Finally, we provide forecasts for future line luminosity function measurements from the JWST and quantify the cosmic variance in such surveys., Comment: 19 pages, 12 figures

The star formation history (SFH) is a key issue in the evolution of galaxies. In this work, we developed a model based on a Gaussian and gamma function mixture to fit SFHs with varying numbers of components. Our primary objective was to use this model to reveal the shape of SFHs and the corresponding physical driving factors. Specifically, we applied this model to fit SFHs from the TNG100-1 simulation. Our study led to the following findings: 1) Our model fits with TNG star formation histories well, especially for high-mass and red galaxies; 2) A clear relationship exists between the number and shape of fitted components and the mass and color of galaxies, with notable differences observed between central/isolated and satellite galaxies. 3) Our model allowed us to extract different episodes of star formation within star formation histories with ease and analyze the duration and timing of each star formation episode. Our findings indicated a strong relationship between the timing of each star formation episode and galaxy mass and color.

As an emerging task that integrates perception and reasoning, topology reasoning in autonomous driving scenes has recently garnered widespread attention. However, existing work often emphasizes "perception over reasoning": they typically boost reasoning performance by enhancing the perception of lanes and directly adopt MLP to learn lane topology from lane query. This paradigm overlooks the geometric features intrinsic to the lanes themselves and are prone to being influenced by inherent endpoint shifts in lane detection. To tackle this issue, we propose an interpretable method for lane topology reasoning based on lane geometric distance and lane query similarity, named TopoLogic. This method mitigates the impact of endpoint shifts in geometric space, and introduces explicit similarity calculation in semantic space as a complement. By integrating results from both spaces, our methods provides more comprehensive information for lane topology. Ultimately, our approach significantly outperforms the existing state-of-the-art methods on the mainstream benchmark OpenLane-V2 (23.9 v.s. 10.9 in TOP$_{ll}$ and 44.1 v.s. 39.8 in OLS on subset_A. Additionally, our proposed geometric distance topology reasoning method can be incorporated into well-trained models without re-training, significantly boost the performance of lane topology reasoning. The code is released at https://github.com/Franpin/TopoLogic.

The analyses of the next generation cosmological surveys demand an accurate, efficient, and differentiable method for simulating the universe and its observables across cosmological volumes. We present Hamiltonian ray tracing (HRT) -- the first post-Born (accounting for lens-lens coupling and without relying on the Born approximation), three-dimensional (without assuming the thin-lens approximation), and on-the-fly (applicable to any structure formation simulations) ray tracing algorithm based on the Hamiltonian formalism. HRT performs symplectic integration of the photon geodesics in a weak gravitational field, and can integrate tightly with any gravity solver, enabling co-evolution of matter particles and light rays with minimal additional computations. We implement HRT in the particle-mesh library $\texttt{pmwd}$, leveraging hardware accelerators such as GPUs and automatic differentiation capabilities based on $\texttt{JAX}$. When tested on a point-mass lens, HRT achieves sub-percent accuracy in deflection angles above the resolution limit across both weak and moderately strong lensing regimes. We also test HRT in cosmological simulations on the convergence maps and their power spectra., Comment: 26 pages, 8 figures; as published in JCAP

Although sparse-view computed tomography (CT) has significantly reduced radiation dose, it also introduces severe artifacts which degrade the image quality. In recent years, deep learning-based methods for inverse problems have made remarkable progress and have become increasingly popular in CT reconstruction. However, most of these methods suffer several limitations: dependence on high-quality training data, weak interpretability, etc. In this study, we propose a fully unsupervised framework called Deep Radon Prior (DRP), inspired by Deep Image Prior (DIP), to address the aforementioned limitations. DRP introduces a neural network as an implicit prior into the iterative method, thereby realizing cross-domain gradient feedback. During the reconstruction process, the neural network is progressively optimized in multiple stages to narrow the solution space in radon domain for the under-constrained imaging protocol, and the convergence of the proposed method has been discussed in this work. Compared with the popular pre-trained method, the proposed framework requires no dataset and exhibits superior interpretability and generalization ability. The experimental results demonstrate that the proposed method can generate detailed images while effectively suppressing image artifacts.Meanwhile, DRP achieves comparable or better performance than the supervised methods., Comment: 11 pages, 12 figures, Journal paper

Recently, integrating video foundation models and large language models to build a video understanding system can overcome the limitations of specific pre-defined vision tasks. Yet, existing systems can only handle videos with very few frames. For long videos, the computation complexity, memory cost, and long-term temporal connection impose additional challenges. Taking advantage of the Atkinson-Shiffrin memory model, with tokens in Transformers being employed as the carriers of memory in combination with our specially designed memory mechanism, we propose the MovieChat to overcome these challenges. MovieChat achieves state-of-the-art performance in long video understanding, along with the released MovieChat-1K benchmark with 1K long video and 14K manual annotations for validation of the effectiveness of our method., Comment: CVPR 2024. First three authors contribute equally to this work. Project Website https://rese1f.github.io/MovieChat/

Nonlinear optical processing of ambient natural light is highly desired in computational imaging and sensing applications. A strong optical nonlinear response that can work under weak broadband incoherent light is essential for this purpose. Here we introduce an optoelectronic nonlinear filter array that can address this emerging need. By merging 2D transparent phototransistors (TPTs) with liquid crystal (LC) modulators, we create an optoelectronic neuron array that allows self-amplitude modulation of spatially incoherent light, achieving a large nonlinear contrast over a broad spectrum at orders-of-magnitude lower intensity than what is achievable in most optical nonlinear materials. For a proof-of-concept demonstration, we fabricated a 10,000-pixel array of optoelectronic neurons, each serving as a nonlinear filter, and experimentally demonstrated an intelligent imaging system that uses the nonlinear response to instantly reduce input glares while retaining the weaker-intensity objects within the field of view of a cellphone camera. This intelligent glare-reduction capability is important for various imaging applications, including autonomous driving, machine vision, and security cameras. Beyond imaging and sensing, this optoelectronic neuron array, with its rapid nonlinear modulation for processing incoherent broadband light, might also find applications in optical computing, where nonlinear activation functions that can work under ambient light conditions are highly sought., Comment: 20 Pages, 5 Figures

3D object detection has become indispensable in the field of autonomous driving. To date, gratifying breakthroughs have been recorded in 3D object detection research, attributed to deep learning. However, deep learning algorithms are data-driven and require large amounts of annotated point cloud data for training and evaluation. Unlike 2D image labels, annotating point cloud data is difficult due to the limitations of sparsity, irregularity, and low resolution, which requires more manual work, and the annotation efficiency is much lower than 2D image.Therefore, we propose an annotation algorithm for point cloud data, which is pre-annotation and camera-LiDAR late fusion algorithm to easily and accurately annotate. The contributions of this study are as follows. We propose (1) a pre-annotation algorithm that employs 3D object detection and auto fitting for the easy annotation of point clouds, (2) a camera-LiDAR late fusion algorithm using 2D and 3D results for easily error checking, which helps annotators easily identify missing objects, and (3) a point cloud annotation evaluation pipeline to evaluate our experiments. The experimental results show that the proposed algorithm improves the annotating speed by 6.5 times and the annotation quality in terms of the 3D Intersection over Union and precision by 8.2 points and 5.6 points, respectively; additionally, the miss rate is reduced by 31.9 points.

Line-intensity mapping (LIM) is a promising technique to constrain the global distribution of galaxy properties. To combine LIM experiments probing different tracers with traditional galaxy surveys and fully exploit the scientific potential of these observations, it is necessary to have a physically motivated modeling framework. As part of developing such a framework, in this work we introduce and model the conditional galaxy property distribution (CGPD), i.e. the distribution of galaxy properties conditioned on the host halo mass and redshift. We consider five galaxy properties, including the galaxy stellar mass, molecular gas mass, galaxy radius, gas phase metallicity and star formation rate (SFR), which are important for predicting the emission lines of interest. The CGPD represents the full distribution of galaxies in the five dimensional property space; many important galaxy distribution functions and scaling relations, such as the stellar mass function and SFR main sequence, can be derived from integrating and projecting it. We utilize two different kinds of cosmological galaxy simulations, a semi-analytic model and the IllustrisTNG hydrodynamic simulation, to characterize the CGPD and explore how well it can be represented using a Gaussian mixture model (GMM). We find that with just a few ($\sim 3$) Gaussian components, a GMM can describe the CGPD of the simulated galaxies to high accuracy for both simulations. The CGPD can be mapped to LIM or other observables by constructing the appropriate relationship between galaxy properties and the relevant observable tracers., Comment: 17 pages, 10 figures

The formation of the large-scale structure, the evolution and distribution of galaxies, quasars, and dark matter on cosmological scales, requires numerical simulations. Differentiable simulations provide gradients of the cosmological parameters, that can accelerate the extraction of physical information from statistical analyses of observational data. The deep learning revolution has brought not only myriad powerful neural networks, but also breakthroughs including automatic differentiation (AD) tools and computational accelerators like GPUs, facilitating forward modeling of the Universe with differentiable simulations. Because AD needs to save the whole forward evolution history to backpropagate gradients, current differentiable cosmological simulations are limited by memory. Using the adjoint method, with reverse time integration to reconstruct the evolution history, we develop a differentiable cosmological particle-mesh (PM) simulation library pmwd (particle-mesh with derivatives) with a low memory cost. Based on the powerful AD library JAX, pmwd is fully differentiable, and is highly performant on GPUs., Comment: repo at https://github.com/eelregit/pmwd

Rapid advances in deep learning have brought not only myriad powerful neural networks, but also breakthroughs that benefit established scientific research. In particular, automatic differentiation (AD) tools and computational accelerators like GPUs have facilitated forward modeling of the Universe with differentiable simulations. Based on analytic or automatic backpropagation, current differentiable cosmological simulations are limited by memory, and thus are subject to a trade-off between time and space/mass resolution, usually sacrificing both. We present a new approach free of such constraints, using the adjoint method and reverse time integration. It enables larger and more accurate forward modeling at the field level, and will improve gradient based optimization and inference. We implement it in an open-source particle-mesh (PM) $N$-body library pmwd (particle-mesh with derivatives). Based on the powerful AD system JAX, pmwd is fully differentiable, and is highly performant on GPUs., Comment: 5 figures + 2 tables; repo at https://github.com/eelregit/pmwd ; v2 matches published version with better typesetting