Song Zhang

Peixiang Huang, Li Liu, Renrui Zhang et al.

To achieve accurate and low-cost 3D object detection, existing methods propose to benefit camera-based multi-view detectors with spatial cues provided by the LiDAR modality, e.g., dense depth supervision and bird-eye-view (BEV) feature distillation. However, they directly conduct point-to-point mimicking from LiDAR to camera, which neglects the inner-geometry of foreground targets and suffers from the modal gap between 2D-3D features. In this paper, we propose the learning scheme of Target Inner-Geometry from the LiDAR modality into camera-based BEV detectors for both dense depth and BEV features, termed as TiG-BEV. First, we introduce an inner-depth supervision module to learn the low-level relative depth relations between different foreground pixels. This enables the camera-based detector to better understand the object-wise spatial structures. Second, we design an inner-feature BEV distillation module to imitate the high-level semantics of different keypoints within foreground targets. To further alleviate the BEV feature gap between two modalities, we adopt both inter-channel and inter-keypoint distillation for feature-similarity modeling. With our target inner-geometry distillation, TiG-BEV can effectively boost BEVDepth by +2.3% NDS and +2.4% mAP, along with BEVDet by +9.1% NDS and +10.3% mAP on nuScenes val set. Code will be available at https://github.com/ADLab3Ds/TiG-BEV.

Eduardo Pérez-Pellitero, Sibi Catley-Chandar, Richard Shaw et al.

This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2022. This manuscript focuses on the competition set-up, datasets, the proposed methods and their results. The challenge aims at estimating an HDR image from multiple respective low dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise. The challenge is composed of two tracks with an emphasis on fidelity and complexity constraints: In Track 1, participants are asked to optimize objective fidelity scores while imposing a low-complexity constraint (i.e. solutions can not exceed a given number of operations). In Track 2, participants are asked to minimize the complexity of their solutions while imposing a constraint on fidelity scores (i.e. solutions are required to obtain a higher fidelity score than the prescribed baseline). Both tracks use the same data and metrics: Fidelity is measured by means of PSNR with respect to a ground-truth HDR image (computed both directly and with a canonical tonemapping operation), while complexity metrics include the number of Multiply-Accumulate (MAC) operations and runtime (in seconds).

Song Zhang, Wenjia Xu, Zhiwei Wei et al.

Multi-View Stereo~(MVS) is a fundamental problem in geometric computer vision which aims to reconstruct a scene using multi-view images with known camera parameters. However, the mainstream approaches represent the scene with a fixed all-pixel depth range and equal depth interval partition, which will result in inadequate utilization of depth planes and imprecise depth estimation. In this paper, we present a novel multi-stage coarse-to-fine framework to achieve adaptive all-pixel depth range and depth interval. We predict a coarse depth map in the first stage, then an Adaptive Depth Range Prediction module is proposed in the second stage to zoom in the scene by leveraging the reference image and the obtained depth map in the first stage and predict a more accurate all-pixel depth range for the following stages. In the third and fourth stages, we propose an Adaptive Depth Interval Adjustment module to achieve adaptive variable interval partition for pixel-wise depth range. The depth interval distribution in this module is normalized by Z-score, which can allocate dense depth hypothesis planes around the potential ground truth depth value and vice versa to achieve more accurate depth estimation. Extensive experiments on four widely used benchmark datasets~(DTU, TnT, BlendedMVS, ETH 3D) demonstrate that our model achieves state-of-the-art performance and yields competitive generalization ability. Particularly, our method achieves the highest Acc and Overall on the DTU dataset, while attaining the highest Recall and $F_{1}$-score on the Tanks and Temples intermediate and advanced dataset. Moreover, our method also achieves the lowest $e_{1}$ and $e_{3}$ on the BlendedMVS dataset and the highest Acc and $F_{1}$-score on the ETH 3D dataset, surpassing all listed methods.Project website: https://github.com/zs670980918/ARAI-MVSNet

Yiyou Sun, Xinyang Han, Weichen Zhang et al.

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 subfields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP-relevant impact.

Song Zhang, Qingzhong Wang, Jiang Bian et al.

While models derived from Vision Transformers (ViTs) have been phonemically surging, pre-trained models cannot seamlessly adapt to arbitrary resolution images without altering the architecture and configuration, such as sampling the positional encoding, limiting their flexibility for various vision tasks. For instance, the Segment Anything Model (SAM) based on ViT-Huge requires all input images to be resized to 1024$\times$1024. To overcome this limitation, we propose the Multi-Head Self-Attention Convolution (MSA-Conv) that incorporates Self-Attention within generalized convolutions, including standard, dilated, and depthwise ones. Enabling transformers to handle images of varying sizes without retraining or rescaling, the use of MSA-Conv further reduces computational costs compared to global attention in ViT, which grows costly as image size increases. Later, we present the Vision Transformer in Convolution (TiC) as a proof of concept for image classification with MSA-Conv, where two capacity enhancing strategies, namely Multi-Directional Cyclic Shifted Mechanism and Inter-Pooling Mechanism, have been proposed, through establishing long-distance connections between tokens and enlarging the effective receptive field. Extensive experiments have been carried out to validate the overall effectiveness of TiC. Additionally, ablation studies confirm the performance improvement made by MSA-Conv and the two capacity enhancing strategies separately. Note that our proposal aims at studying an alternative to the global attention used in ViT, while MSA-Conv meets our goal by making TiC comparable to state-of-the-art on ImageNet-1K. Code will be released at https://github.com/zs670980918/MSA-Conv.

Qingsen Yan, Song Zhang, Weiye Chen et al.

Generating a high-quality High Dynamic Range (HDR) image from dynamic scenes has recently been extensively studied by exploiting Deep Neural Networks (DNNs). Most DNNs-based methods require a large amount of training data with ground truth, requiring tedious and time-consuming work. Few-shot HDR imaging aims to generate satisfactory images with limited data. However, it is difficult for modern DNNs to avoid overfitting when trained on only a few images. In this work, we propose a novel semi-supervised approach to realize few-shot HDR imaging via two stages of training, called SSHDR. Unlikely previous methods, directly recovering content and removing ghosts simultaneously, which is hard to achieve optimum, we first generate content of saturated regions with a self-supervised mechanism and then address ghosts via an iterative semi-supervised learning framework. Concretely, considering that saturated regions can be regarded as masking Low Dynamic Range (LDR) input regions, we design a Saturated Mask AutoEncoder (SMAE) to learn a robust feature representation and reconstruct a non-saturated HDR image. We also propose an adaptive pseudo-label selection strategy to pick high-quality HDR pseudo-labels in the second stage to avoid the effect of mislabeled samples. Experiments demonstrate that SSHDR outperforms state-of-the-art methods quantitatively and qualitatively within and across different datasets, achieving appealing HDR visualization with few labeled samples.

Qingsen Yan, Weiye Chen, Song Zhang et al.

Mapping Low Dynamic Range (LDR) images with different exposures to High Dynamic Range (HDR) remains nontrivial and challenging on dynamic scenes due to ghosting caused by object motion or camera jitting. With the success of Deep Neural Networks (DNNs), several DNNs-based methods have been proposed to alleviate ghosting, they cannot generate approving results when motion and saturation occur. To generate visually pleasing HDR images in various cases, we propose a hybrid HDR deghosting network, called HyHDRNet, to learn the complicated relationship between reference and non-reference images. The proposed HyHDRNet consists of a content alignment subnetwork and a Transformer-based fusion subnetwork. Specifically, to effectively avoid ghosting from the source, the content alignment subnetwork uses patch aggregation and ghost attention to integrate similar content from other non-reference images with patch level and suppress undesired components with pixel level. To achieve mutual guidance between patch-level and pixel-level, we leverage a gating module to sufficiently swap useful information both in ghosted and saturated regions. Furthermore, to obtain a high-quality HDR image, the Transformer-based fusion subnetwork uses a Residual Deformable Transformer Block (RDTB) to adaptively merge information for different exposed regions. We examined the proposed method on four widely used public HDR image deghosting datasets. Experiments demonstrate that HyHDRNet outperforms state-of-the-art methods both quantitatively and qualitatively, achieving appealing HDR visualization with unified textures and colors.

Jinjin Gu, Jinan Zhou, Ringo Sai Wo Chu et al.

Event cameras are novel bio-inspired vision sensors that output pixel-level intensity changes in microsecond accuracy with a high dynamic range and low power consumption. Despite these advantages, event cameras cannot be directly applied to computational imaging tasks due to the inability to obtain high-quality intensity and events simultaneously. This paper aims to connect a standalone event camera and a modern intensity camera so that the applications can take advantage of both two sensors. We establish this connection through a multi-modal stereo matching task. We first convert events to a reconstructed image and extend the existing stereo networks to this multi-modality condition. We propose a self-supervised method to train the multi-modal stereo network without using ground truth disparity data. The structure loss calculated on image gradients is used to enable self-supervised learning on such multi-modal data. Exploiting the internal stereo constraint between views with different modalities, we introduce general stereo loss functions, including disparity cross-consistency loss and internal disparity loss, leading to improved performance and robustness compared to existing approaches. The experiments demonstrate the effectiveness of the proposed method, especially the proposed general stereo loss functions, on both synthetic and real datasets. At last, we shed light on employing the aligned events and intensity images in downstream tasks, e.g., video interpolation application.

Song Zhang, Yuqing Duan, Daoliang Li et al.

In underwater image enhancement (UIE), convolutional neural networks (CNN) have inherent limitations in modeling long-range dependencies and are less effective in recovering global features. While Transformers excel at modeling long-range dependencies, their quadratic computational complexity with increasing image resolution presents significant efficiency challenges. Additionally, most supervised learning methods lack effective physical model constraint, which can lead to insufficient realism and overfitting in generated images. To address these issues, we propose a physical model constraint-based underwater image enhancement framework, Mamba-UIE. Specifically, we decompose the input image into four components: underwater scene radiance, direct transmission map, backscatter transmission map, and global background light. These components are reassembled according to the revised underwater image formation model, and the reconstruction consistency constraint is applied between the reconstructed image and the original image, thereby achieving effective physical constraint on the underwater image enhancement process. To tackle the quadratic computational complexity of Transformers when handling long sequences, we introduce the Mamba-UIE network based on linear complexity state space models. By incorporating the Mamba in Convolution block, long-range dependencies are modeled at both the channel and spatial levels, while the CNN backbone is retained to recover local features and details. Extensive experiments on three public datasets demonstrate that our proposed Mamba-UIE outperforms existing state-of-the-art methods, achieving a PSNR of 27.13 and an SSIM of 0.93 on the UIEB dataset. Our method is available at https://github.com/zhangsong1213/Mamba-UIE.

Song Zhang, Daoliang Li, Ran Zhao

Majority of deep learning methods utilize vanilla convolution for enhancing underwater images. While vanilla convolution excels in capturing local features and learning the spatial hierarchical structure of images, it tends to smooth input images, which can somewhat limit feature expression and modeling. A prominent characteristic of underwater degraded images is blur, and the goal of enhancement is to make the textures and details (high-frequency features) in the images more visible. Therefore, we believe that leveraging high-frequency features can improve enhancement performance. To address this, we introduce Pixel Difference Convolution (PDC), which focuses on gradient information with significant changes in the image, thereby improving the modeling of enhanced images. We propose an underwater image enhancement network, PDCFNet, based on PDC and cross-level feature fusion. Specifically, we design a detail enhancement module based on PDC that employs parallel PDCs to capture high-frequency features, leading to better detail and texture enhancement. The designed cross-level feature fusion module performs operations such as concatenation and multiplication on features from different levels, ensuring sufficient interaction and enhancement between diverse features. Our proposed PDCFNet achieves a PSNR of 27.37 and an SSIM of 92.02 on the UIEB dataset, attaining the best performance to date. Our code is available at https://github.com/zhangsong1213/PDCFNet.

Zibo Zhao, Zeqiang Lai, Qingxiang Lin et al.

We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scale shape generation model -- Hunyuan3D-DiT, and a large-scale texture synthesis model -- Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we build Hunyuan3D-Studio -- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2

Hongyi Lin, Wenxiu Shi, Heye Huang et al.

Generating safety-critical driving scenarios is crucial for evaluating and improving autonomous driving systems, but long-tail risky situations are rarely observed in real-world data and difficult to specify through manual scenario design. Existing generative approaches typically treat risk as an after-the-fact label and struggle to maintain geometric consistency in multi-view driving scenes. We present RiskMV-DPO, a general and systematic pipeline for physically-informed, risk-controllable multi-view scenario generation. By integrating target risk levels with physically-grounded risk modeling, we autonomously synthesize diverse and high-stakes dynamic trajectories that serve as explicit geometric anchors for a diffusion-based video generator. To ensure spatial-temporal coherence and geometric fidelity, we introduce a geometry-appearance alignment module and a region-aware direct preference optimization (RA-DPO) strategy with motion-aware masking to focus learning on localized dynamic regions.Experiments on the nuScenes dataset show that RiskMV-DPO can freely generate a wide spectrum of diverse long-tail scenarios while maintaining state-of-the-art visual quality, improving 3D detection mAP from 18.17 to 30.50 and reducing FID to 15.70. Our work shifts the role of world models from passive environment prediction to proactive, risk-controllable synthesis, providing a scalable toolchain for the safety-oriented development of embodied intelligence.

Song Zhang, Ruohan Guo, Xiaohua Ge et al.

Reliable health assessment of retired lithium-ion batteries is essential for safe and economically viable second-life deployment, yet remains difficult due to sparse measurements, incomplete historical records, heterogeneous chemistries, and limited or noisy battery health labels. Conventional laboratory diagnostics, such as full charge-discharge cycling, pulse tests, Electrochemical Impedance Spectroscopy (EIS) measurements, and thermal characterization, provide accurate degradation information but are too time-consuming, equipment-intensive, or condition-sensitive to be applied at scale during retirement-stage sorting, leaving real-world datasets fragmented and inconsistent. This review synthesizes recent advances that address these constraints through physical health indicators, experiment testing methods, data-generation and augmentation techniques, and a spectrum of learning-based modeling routes spanning supervised, semi-supervised, weakly supervised, and unsupervised paradigms. We highlight how minimal-test features, synthetic data, domain-invariant representations, and uncertainty-aware prediction enable robust inference under limited or approximate labels and across mixed chemistries and operating histories. A comparative evaluation further reveals trade-offs in accuracy, interpretability, scalability, and computational burden. Looking forward, progress toward physically constrained generative models, cross-chemistry generalization, calibrated uncertainty estimation, and standardized benchmarks will be crucial for building reliable, scalable, and deployment-ready health prediction tools tailored to the realities of retired-battery applications.

Song Zhang, Ken Zheng, Xiaoxu Zhu et al.

Grapheme-to-phoneme (G2P) conversion is an indispensable part of the Chinese Mandarin text-to-speech (TTS) system, and the core of G2P conversion is to solve the problem of polyphone disambiguation, which is to pick up the correct pronunciation for several candidates for a Chinese polyphonic character. In this paper, we propose a Chinese polyphone BERT model to predict the pronunciations of Chinese polyphonic characters. Firstly, we create 741 new Chinese monophonic characters from 354 source Chinese polyphonic characters by pronunciation. Then we get a Chinese polyphone BERT by extending a pre-trained Chinese BERT with 741 new Chinese monophonic characters and adding a corresponding embedding layer for new tokens, which is initialized by the embeddings of source Chinese polyphonic characters. In this way, we can turn the polyphone disambiguation task into a pre-training task of the Chinese polyphone BERT. Experimental results demonstrate the effectiveness of the proposed model, and the polyphone BERT model obtain 2% (from 92.1% to 94.1%) improvement of average accuracy compared with the BERT-based classifier model, which is the prior state-of-the-art in polyphone disambiguation.

Yixiong Huo, Guangfeng Jiang, Hongyang Wei et al.

3D Gaussian Splatting (3D GS) has gained popularity due to its faster rendering speed and high-quality novel view synthesis. Some researchers have explored using 3D GS for reconstructing driving scenes. However, these methods often rely on various data types, such as depth maps, 3D boxes, and trajectories of moving objects. Additionally, the lack of annotations for synthesized images limits their direct application in downstream tasks. To address these issues, we propose EGSRAL, a 3D GS-based method that relies solely on training images without extra annotations. EGSRAL enhances 3D GS's capability to model both dynamic objects and static backgrounds and introduces a novel adaptor for auto labeling, generating corresponding annotations based on existing annotations. We also propose a grouping strategy for vanilla 3D GS to address perspective issues in rendering large-scale, complex scenes. Our method achieves state-of-the-art performance on multiple datasets without any extra annotation. For example, the PSNR metric reaches 29.04 on the nuScenes dataset. Moreover, our automated labeling can significantly improve the performance of 2D/3D detection tasks. Code is available at https://github.com/jiangxb98/EGSRAL.

Song Zhang, Qingzhong Wang, Junyi Liu et al.

In the fast-growing field of Remote Sensing (RS) image analysis, the gap between massive unlabeled datasets and the ability to fully utilize these datasets for advanced RS analytics presents a significant challenge. To fill the gap, our work introduces an innovative auto-labeling framework named ALPS (Automatic Labeling for Pre-training in Segmentation), leveraging the Segment Anything Model (SAM) to predict precise pseudo-labels for RS images without necessitating prior annotations or additional prompts. The proposed pipeline significantly reduces the labor and resource demands traditionally associated with annotating RS datasets. By constructing two comprehensive pseudo-labeled RS datasets via ALPS for pre-training purposes, our approach enhances the performance of downstream tasks across various benchmarks, including iSAID and ISPRS Potsdam. Experiments demonstrate the effectiveness of our framework, showcasing its ability to generalize well across multiple tasks even under the scarcity of extensively annotated datasets, offering a scalable solution to automatic segmentation and annotation challenges in the field. In addition, the proposed a pipeline is flexible and can be applied to medical image segmentation, remarkably boosting the performance. Note that ALPS utilizes pre-trained SAM to semi-automatically annotate RS images without additional manual annotations. Though every component in the pipeline has bee well explored, integrating clustering algorithms with SAM and novel pseudo-label alignment significantly enhances RS segmentation, as an off-the-shelf tool for pre-training data preparation. Our source code is available at: https://github.com/StriveZs/ALPS.

Song Zhang, Yanlong Chen, Yilin Li et al.

Remote sensing vision-language models (RS-VLMs) face a fundamental mismatch with natural-image counterparts: the same geographic object exhibits radically different visual evidence across ground sampling distances (GSDs) spanning multiple orders of magnitude. Yet existing RS-VLMs often discard GSD or inject it as a discrete text token, forcing a single static parameter set to absorb the entire scale spectrum. We introduce ScaleEarth, a parameter-efficient fine-tuning framework built on Qwen3-VL that treats GSD as a continuous conditioning variable governing the model's computation path. At its core, CS-HLoRA (Continuous Scale-Conditioned Hyper-LoRA) modulates the LoRA low-rank subspace through a GSD-driven gate, enabling the model to dynamically route computation by physical scale. To remove reliance on sensor metadata at deployment, we pair CS-HLoRA with SSE-U, a lightweight heteroscedastic sub-head that predicts GSD and its uncertainty from visual features. To provide matching supervision, we construct GeoScale-VQA, a 1.5M-sample scale-layered RS-VQA corpus whose question-answer generation is conditioned on the same physical scalar that drives CS-HLoRA, forming a closed method-data loop. Trained with QLoRA on an 8B backbone, ScaleEarth achieves state-of-the-art results on remote-sensing benchmarks covering diverse Earth-system tasks, including XLRS-Bench and OmniEarth-Bench.

Ruibo Wang, Song Zhang, Ping Huang et al.

Recent research that combines implicit 3D representation with semantic information, like Semantic-NeRF, has proven that NeRF model could perform excellently in rendering 3D structures with semantic labels. This research aims to extend the Semantic Neural Radiance Fields (Semantic-NeRF) model by focusing solely on semantic output and removing the RGB output component. We reformulate the model and its training procedure to leverage only the cross-entropy loss between the model semantic output and the ground truth semantic images, removing the colour data traditionally used in the original Semantic-NeRF approach. We then conduct a series of identical experiments using the original and the modified Semantic-NeRF model. Our primary objective is to obverse the impact of this modification on the model performance by Semantic-NeRF, focusing on tasks such as scene understanding, object detection, and segmentation. The results offer valuable insights into the new way of rendering the scenes and provide an avenue for further research and development in semantic-focused 3D scene understanding.

Ruibo Wang, Song Zhang, Ping Huang et al.

Accurately reconstructing road surfaces is pivotal for various applications especially in autonomous driving. This paper introduces a position encoding Multi-Layer Perceptrons (MLPs) framework to reconstruct road surfaces, with input as world coordinates x and y, and output as height, color, and semantic information. The effectiveness of this method is demonstrated through its compatibility with a variety of road height sources like vehicle camera poses, LiDAR point clouds, and SFM point clouds, robust to the semantic noise of images like sparse labels and noise semantic prediction, and fast training speed, which indicates a promising application for rendering road surfaces with semantics, particularly in applications demanding visualization of road surface, 4D labeling, and semantic groupings.

Song Zhang, Zhiwei Wei, Wenjia Xu et al.

3D terrain reconstruction with remote sensing imagery achieves cost-effective and large-scale earth observation and is crucial for safeguarding natural disasters, monitoring ecological changes, and preserving the environment.Recently, learning-based multi-view stereo~(MVS) methods have shown promise in this task. However, these methods simply modify the general learning-based MVS framework for height estimation, which overlooks the terrain characteristics and results in insufficient accuracy. Considering that the Earth's surface generally undulates with no drastic changes and can be measured by slope, integrating slope considerations into MVS frameworks could enhance the accuracy of terrain reconstructions. To this end, we propose an end-to-end slope-aware height estimation network named TS-SatMVSNet for large-scale remote sensing terrain reconstruction.To effectively obtain the slope representation, drawing from mathematical gradient concepts, we innovatively proposed a height-based slope calculation strategy to first calculate a slope map from a height map to measure the terrain undulation. To fully integrate slope information into the MVS pipeline, we separately design two slope-guided modules to enhance reconstruction outcomes at both micro and macro levels. Specifically, at the micro level, we designed a slope-guided interval partition module for refined height estimation using slope values. At the macro level, a height correction module is proposed, using a learnable Gaussian smoothing operator to amend the inaccurate height values. Additionally, to enhance the efficacy of height estimation, we proposed a slope direction loss for implicitly optimizing height estimation results. Extensive experiments on the WHU-TLC dataset and MVS3D dataset show that our proposed method achieves state-of-the-art performance and demonstrates competitive generalization ability.

Song Zhang, Haoyu Chen, Ruibo Wang

Humans combine prediction and perception to observe the world. When faced with rapidly moving birds or insects, we can only perceive them clearly by predicting their next position and focusing our gaze there. Inspired by this, this paper proposes the Prediction-As-Perception (PAP) framework, integrating a prediction-perception architecture into 3D object perception tasks to enhance the model's perceptual accuracy. The PAP framework consists of two main modules: prediction and perception, primarily utilizing continuous frame information as input. Firstly, the prediction module forecasts the potential future positions of ego vehicles and surrounding traffic participants based on the perception results of the current frame. These predicted positions are then passed as queries to the perception module of the subsequent frame. The perceived results are iteratively fed back into the prediction module. We evaluated the PAP structure using the end-to-end model UniAD on the nuScenes dataset. The results demonstrate that the PAP structure improves UniAD's target tracking accuracy by 10% and increases the inference speed by 15%. This indicates that such a biomimetic design significantly enhances the efficiency and accuracy of perception models while reducing computational resource consumption.

Raúl Vargas, Lenny A. Romero, Song Zhang et al.

This Letter presents a novel structured light system model that effectively considers local lens distortion by pixel-wise rational functions. We leverage the stereo method for initial calibration and then estimate the rational model for each pixel. Our proposed model can achieve high measurement accuracy within and outside the calibration volume, demonstrating its robustness and accuracy.

Song Zhang, Xiaochuan Luo, Eugene Litvinov

Operating a modern power grid reliably in case of SCADA/EMS failure or amid difficult times like COVID-19 pandemic is a challenging task for grid operators. In [11], a PMU-based emergency generation dispatch scheme has been proposed to help the system operators with the supply and demand balancing; however, its realization highly relies on the control center infrastructure for computing and communication. This work, rather than using the on-premises server and dispatch communication system, proposes and implements a cloud-centric serverless architecture to ensure the operation continuity regardless of local infrastructure's availability and accessibility. Through its prototype implementation and evaluation at ISO New England, the solution has demonstrated two major advantages. Firstly, the cloud infrastructure is independent and fault-tolerant, providing grid monitoring and control capability even when EMS loses the corresponding functionality or when operators need to work remotely away from the control center. Secondly, the overall design is event-driven using serverless cloud services in response to the SCADA/EMS failure event. Thanks to "serverless", the burden of the server provisioning and maintenance can be avoided from the user side. The cost of using public cloud services for this solution is extremely low since it is architected and implemented based on the event-driven Function-as-a-Service (FaaS) model. This work also develops a comprehensive cyber security mechanism to comply with critical infrastructure requirements for the power grid, which can serve as an exemplary framework for other grid operators to secure their cloud services.

Song Zhang, Gehui Shen, Jinsong Huang et al.

Lifelong or continual learning remains to be a challenge for artificial neural network, as it is required to be both stable for preservation of old knowledge and plastic for acquisition of new knowledge. It is common to see previous experience get overwritten, which leads to the well-known issue of catastrophic forgetting, especially in the scenario of class-incremental learning (Class-IL). Recently, many lifelong learning methods have been proposed to avoid catastrophic forgetting. However, models which learn without replay of the input data, would encounter another problem which has been ignored, and we refer to it as prior information loss (PIL). In training procedure of Class-IL, as the model has no knowledge about following tasks, it would only extract features necessary for tasks learned so far, whose information is insufficient for joint classification. In this paper, our empirical results on several image datasets show that PIL limits the performance of current state-of-the-art method for Class-IL, the orthogonal weights modification (OWM) algorithm. Furthermore, we propose to combine self-supervised learning, which can provide effective representations without requiring labels, with Class-IL to partly get around this problem. Experiments show superiority of proposed method to OWM, as well as other strong baselines.

Gehui Shen, Song Zhang, Xiang Chen et al.

The ability of intelligent agents to learn and remember multiple tasks sequentially is crucial to achieving artificial general intelligence. Many continual learning (CL) methods have been proposed to overcome catastrophic forgetting which results from non i.i.d data in the sequential learning of neural networks. In this paper we focus on class incremental learning, a challenging CL scenario. For this scenario, generative replay is a promising strategy which generates and replays pseudo data for previous tasks to alleviate catastrophic forgetting. However, it is hard to train a generative model continually for relatively complex data. Based on recently proposed orthogonal weight modification (OWM) algorithm which can approximately keep previously learned feature invariant when learning new tasks, we propose to 1) replay penultimate layer feature with a generative model; 2) leverage a self-supervised auxiliary task to further enhance the stability of feature. Empirical results on several datasets show our method always achieves substantial improvement over powerful OWM while conventional generative replay always results in a negative effect. Meanwhile our method beats several strong baselines including one based on real data storage. In addition, we conduct experiments to study why our method is effective.

Xinting Yang, Song Zhang, Jintao Liu et al.

With the rapid emergence of deep learning (DL) technology, it has been successfully used in various fields including aquaculture. This change can create new opportunities and a series of challenges for information and data processing in smart fish farming. This paper focuses on the applications of DL in aquaculture, including live fish identification, species classification, behavioral analysis, feeding decision-making, size or biomass estimation, water quality prediction. In addition, the technical details of DL methods applied to smart fish farming are also analyzed, including data, algorithms, computing power, and performance. The results of this review show that the most significant contribution of DL is the ability to automatically extract features. However, challenges still exist; DL is still in an era of weak artificial intelligence. A large number of labeled data are needed for training, which has become a bottleneck restricting further DL applications in aquaculture. Nevertheless, DL still offers breakthroughs in the handling of complex data in aquaculture. In brief, our purpose is to provide researchers and practitioners with a better understanding of the current state of the art of DL in aquaculture, which can provide strong support for the implementation of smart fish farming.

Raul Vargas, Andres G. Marrugo, Song Zhang et al.

The key to accurate 3D shape measurement in Fringe Projection Profilometry (FPP) is the proper calibration of the measurement system. Current calibration techniques rely on phase-coordinate mapping (PCM) or back-projection stereo-vision (SV) methods. PCM methods are cumbersome to implement as they require precise positioning of the calibration target relative to the FPP system but produce highly accurate measurements within the calibration volume. SV methods generally do not achieve the same accuracy level. However, the calibration is more flexible in that the calibration target can be arbitrarily positioned. In this work, we propose a hybrid calibration method that leverages the SV calibration approach using a PCM method to achieve higher accuracy. The method has the flexibility of SV methods, is robust to lens distortions, and has a simple relation between the recovered phase and the metric coordinates. Experimental results show that the proposed Hybrid method outperforms the SV method in terms of accuracy and reconstruction time due to its low computational complexity.