Zhenbo Song

Zhaoxin Fan, Kaixing Yang, Min Zhang et al.

Recently, over-height vehicle strike frequently occurs, causing great economic cost and serious safety problems. Hence, an alert system which can accurately discover any possible height limiting devices in advance is necessary to be employed in modern large or medium sized cars, such as touring cars. Detecting and estimating the height limiting devices act as the key point of a successful height limit alert system. Though there are some works research height limit estimation, existing methods are either too computational expensive or not accurate enough. In this paper, we propose a novel stereo-based pipeline named SHLE for height limit estimation. Our SHLE pipeline consists of two stages. In stage 1, a novel devices detection and tracking scheme is introduced, which accurately locate the height limit devices in the left or right image. Then, in stage 2, the depth is temporally measured, extracted and filtered to calculate the height limit device. To benchmark the height limit estimation task, we build a large-scale dataset named "Disparity Height", where stereo images, pre-computed disparities and ground-truth height limit annotations are provided. We conducted extensive experiments on "Disparity Height" and the results show that SHLE achieves an average error below than 10cm though the car is 70m away from the devices. Our method also outperforms all compared baselines and achieves state-of-the-art performance. Code is available at https://github.com/Yang-Kaixing/SHLE.

Zhenyuan Zhang, Zhenbo Song, Kaihao Zhang et al.

Deep learning based methods have achieved significant success in the task of single image reflection removal (SIRR). However, the majority of these methods are focused on High-Definition/Standard-Definition (HD/SD) images, while ignoring higher resolution images such as Ultra-High-Definition (UHD) images. With the increasing prevalence of UHD images captured by modern devices, in this paper, we aim to address the problem of UHD SIRR. Specifically, we first synthesize two large-scale UHD datasets, UHDRR4K and UHDRR8K. The UHDRR4K dataset consists of $2,999$ and $168$ quadruplets of images for training and testing respectively, and the UHDRR8K dataset contains $1,014$ and $105$ quadruplets. To the best of our knowledge, these two datasets are the first largest-scale UHD datasets for SIRR. Then, we conduct a comprehensive evaluation of six state-of-the-art SIRR methods using the proposed datasets. Based on the results, we provide detailed discussions regarding the strengths and limitations of these methods when applied to UHD images. Finally, we present a transformer-based architecture named RRFormer for reflection removal. RRFormer comprises three modules, namely the Prepossessing Embedding Module, Self-attention Feature Extraction Module, and Multi-scale Spatial Feature Extraction Module. These modules extract hypercolumn features, global and partial attention features, and multi-scale spatial features, respectively. To ensure effective training, we utilize three terms in our loss function: pixel loss, feature loss, and adversarial loss. We demonstrate through experimental results that RRFormer achieves state-of-the-art performance on both the non-UHD dataset and our proposed UHDRR datasets. The code and datasets are publicly available at https://github.com/Liar-zzy/Benchmarking-Ultra-High-Definition-Single-Image-Reflection-Removal.

Han Sun, Zhaoxin Fan, Zhenbo Song et al.

Monocular 3D object detection is a fundamental but very important task to many applications including autonomous driving, robotic grasping and augmented reality. Existing leading methods tend to estimate the depth of the input image first, and detect the 3D object based on point cloud. This routine suffers from the inherent gap between depth estimation and object detection. Besides, the prediction error accumulation would also affect the performance. In this paper, a novel method named MonoSIM is proposed. The insight behind introducing MonoSIM is that we propose to simulate the feature learning behaviors of a point cloud based detector for monocular detector during the training period. Hence, during inference period, the learned features and prediction would be similar to the point cloud based detector as possible. To achieve it, we propose one scene-level simulation module, one RoI-level simulation module and one response-level simulation module, which are progressively used for the detector's full feature learning and prediction pipeline. We apply our method to the famous M3D-RPN detector and CaDDN detector, conducting extensive experiments on KITTI and Waymo Open datasets. Results show that our method consistently improves the performance of different monocular detectors for a large margin without changing their network architectures. Our codes will be publicly available at https://github.com/sunh18/MonoSIM}{https://github.com/sunh18/MonoSIM.

Ziqiao Peng, Haoyu Wu, Zhenbo Song et al.

Speech-driven 3D face animation aims to generate realistic facial expressions that match the speech content and emotion. However, existing methods often neglect emotional facial expressions or fail to disentangle them from speech content. To address this issue, this paper proposes an end-to-end neural network to disentangle different emotions in speech so as to generate rich 3D facial expressions. Specifically, we introduce the emotion disentangling encoder (EDE) to disentangle the emotion and content in the speech by cross-reconstructed speech signals with different emotion labels. Then an emotion-guided feature fusion decoder is employed to generate a 3D talking face with enhanced emotion. The decoder is driven by the disentangled identity, emotional, and content embeddings so as to generate controllable personal and emotional styles. Finally, considering the scarcity of the 3D emotional talking face data, we resort to the supervision of facial blendshapes, which enables the reconstruction of plausible 3D faces from 2D emotional data, and contribute a large-scale 3D emotional talking face dataset (3D-ETF) to train the network. Our experiments and user studies demonstrate that our approach outperforms state-of-the-art methods and exhibits more diverse facial movements. We recommend watching the supplementary video: https://ziqiaopeng.github.io/emotalk

Zhaoxin Fan, Zhenbo Song, Jian Xu et al.

Recently, RGBD-based category-level 6D object pose estimation has achieved promising improvement in performance, however, the requirement of depth information prohibits broader applications. In order to relieve this problem, this paper proposes a novel approach named Object Level Depth reconstruction Network (OLD-Net) taking only RGB images as input for category-level 6D object pose estimation. We propose to directly predict object-level depth from a monocular RGB image by deforming the category-level shape prior into object-level depth and the canonical NOCS representation. Two novel modules named Normalized Global Position Hints (NGPH) and Shape-aware Decoupled Depth Reconstruction (SDDR) module are introduced to learn high fidelity object-level depth and delicate shape representations. At last, the 6D object pose is solved by aligning the predicted canonical representation with the back-projected object-level depth. Extensive experiments on the challenging CAMERA25 and REAL275 datasets indicate that our model, though simple, achieves state-of-the-art performance.

Zhenbo Song, Xianghui Ze, Jianfeng Lu et al.

This paper addresses the problem of estimating the 3-DoF camera pose for a ground-level image with respect to a satellite image that encompasses the local surroundings. We propose a novel end-to-end approach that leverages the learning of dense pixel-wise flow fields in pairs of ground and satellite images to calculate the camera pose. Our approach differs from existing methods by constructing the feature metric at the pixel level, enabling full-image supervision for learning distinctive geometric configurations and visual appearances across views. Specifically, our method employs two distinct convolution networks for ground and satellite feature extraction. Then, we project the ground feature map to the bird's eye view (BEV) using a fixed camera height assumption to achieve preliminary geometric alignment. To further establish content association between the BEV and satellite features, we introduce a residual convolution block to refine the projected BEV feature. Optical flow estimation is performed on the refined BEV feature map and the satellite feature map using flow decoder networks based on RAFT. After obtaining dense flow correspondences, we apply the least square method to filter matching inliers and regress the ground camera pose. Extensive experiments demonstrate significant improvements compared to state-of-the-art methods. Notably, our approach reduces the median localization error by 89%, 19%, 80% and 35% on the KITTI, Ford multi-AV, VIGOR and Oxford RobotCar datasets, respectively.

Zhaoxin Fan, Zhenbo Song, Hongyan Liu et al.

Large-scale place recognition is a fundamental but challenging task, which plays an increasingly important role in autonomous driving and robotics. Existing methods have achieved acceptable good performance, however, most of them are concentrating on designing elaborate global descriptor learning network structures. The importance of feature generalization and descriptor post-enhancing has long been neglected. In this work, we propose a novel method named GIDP to learn a Good Initialization and Inducing Descriptor Poseenhancing for Large-scale Place Recognition. In particular, an unsupervised momentum contrast point cloud pretraining module and a reranking-based descriptor post-enhancing module are proposed respectively in GIDP. The former aims at learning a good initialization for the point cloud encoding network before training the place recognition model, while the later aims at post-enhancing the predicted global descriptor through reranking at inference time. Extensive experiments on both indoor and outdoor datasets demonstrate that our method can achieve state-of-the-art performance using simple and general point cloud encoding backbones.

Zhaoxin Fan, Yuqing Pan, Hao Xu et al.

In the field of full-body reconstruction, the scarcity of annotated data often impedes the efficacy of prevailing methods. To address this issue, we introduce FuRPE, a novel framework that employs part-experts and an ingenious pseudo ground-truth selection scheme to derive high-quality pseudo labels. These labels, central to our approach, equip our network with the capability to efficiently learn from the available data. Integral to FuRPE is a unique exponential moving average training strategy and expert-derived feature distillation strategy. These novel elements of FuRPE not only serve to further refine the model but also to reduce potential biases that may arise from inaccuracies in pseudo labels, thereby optimizing the network's training process and enhancing the robustness of the model. We apply FuRPE to train both two-stage and fully convolutional single-stage full-body reconstruction networks. Our exhaustive experiments on numerous benchmark datasets illustrate a substantial performance boost over existing methods, underscoring FuRPE's potential to reshape the state-of-the-art in full-body reconstruction.

Xukun Zhou, Zhenbo Song, Jun He et al.

Scene Graph Generation is a critical enabler of environmental comprehension for autonomous robotic systems. Most of existing methods, however, are often thwarted by the intricate dynamics of background complexity, which limits their ability to fully decode the inherent topological information of the environment. Additionally, the wealth of contextual information encapsulated within depth cues is often left untapped, rendering existing approaches less effective. To address these shortcomings, we present STDG, an avant-garde Depth-Guided One-Stage Scene Graph Generation methodology. The innovative architecture of STDG is a triad of custom-built modules: The Depth Guided HHA Representation Generation Module, the Depth Guided Semi-Teaching Network Learning Module, and the Depth Guided Scene Graph Generation Module. This trifecta of modules synergistically harnesses depth information, covering all aspects from depth signal generation and depth feature utilization, to the final scene graph prediction. Importantly, this is achieved without imposing additional computational burden during the inference phase. Experimental results confirm that our method significantly enhances the performance of one-stage scene graph generation baselines.

Tianbao Zhang, Zhenyu Liang, Zhenbo Song et al.

High-resolution geometric prediction is essential for robust perception in autonomous driving, robotics, and AR/MR, but current foundation models are fundamentally limited by their scalability to real-world, high-resolution scenarios. Direct inference on 2K images with these models incurs prohibitive computational and memory demands, making practical deployment challenging. To tackle the issue, we present 2K Retrofit, a novel framework that enables efficient 2K-resolution inference for any geometric foundation model, without modifying or retraining the backbone. Our approach leverages fast coarse predictions and an entropy-based sparse refinement to selectively enhance high-uncertainty regions, achieving precise and high-fidelity 2K outputs with minimal overhead. Extensive experiments on widely used benchmark demonstrate that 2K Retrofit consistently achieves state-of-the-art accuracy and speed, bridging the gap between research advances and scalable deployment in high-resolution 3D vision applications. Code will be released upon acceptance.

Haoyun Li, Ivan Zhang, Runqi Ouyang et al.

Vision Language Action (VLA) models derive their generalization capability from diverse training data, yet collecting embodied robot interaction data remains prohibitively expensive. In contrast, human demonstration videos are far more scalable and cost-efficient to collect, and recent studies confirm their effectiveness in training VLA models. However, a significant domain gap persists between human videos and robot-executed videos, including unstable camera viewpoints, visual discrepancies between human hands and robotic arms, and differences in motion dynamics. To bridge this gap, we propose MimicDreamer, a framework that turns fast, low-cost human demonstrations into robot-usable supervision by jointly aligning vision, viewpoint, and actions to directly support policy training. For visual alignment, we propose H2R Aligner, a video diffusion model that generates high-fidelity robot demonstration videos by transferring motion from human manipulation footage. For viewpoint stabilization, EgoStabilizer is proposed, which canonicalizes egocentric videos via homography and inpaints occlusions and distortions caused by warping. For action alignment, we map human hand trajectories to the robot frame and apply a constrained inverse kinematics solver to produce feasible, low-jitter joint commands with accurate pose tracking. Empirically, VLA models trained purely on our synthesized human-to-robot videos achieve few-shot execution on real robots. Moreover, scaling training with human data significantly boosts performance compared to models trained solely on real robot data; our approach improves the average success rate by 14.7\% across six representative manipulation tasks.

Junhao Wu, Xiuer Gu, Zhiying Li et al.

Evaluating human actions with clear and detailed feedback is important in areas such as sports, healthcare, and robotics, where decisions rely not only on final outcomes but also on interpretable reasoning. However, most existing methods provide only a final score without explanation or detailed analysis, limiting their practical applicability. To address this, we introduce HieroAction, a vision-language model that delivers accurate and structured assessments of human actions. HieroAction builds on two key ideas: (1) Stepwise Action Reasoning, a tailored chain of thought process designed specifically for action assessment, which guides the model to evaluate actions step by step, from overall recognition through sub action analysis to final scoring, thus enhancing interpretability and structured understanding; and (2) Hierarchical Policy Learning, a reinforcement learning strategy that enables the model to learn fine grained sub action dynamics and align them with high level action quality, thereby improving scoring precision. The reasoning pathway structures the evaluation process, while policy learning refines each stage through reward based optimization. Their integration ensures accurate and interpretable assessments, as demonstrated by superior performance across multiple benchmark datasets. Code will be released upon acceptance.

Xianghui Ze, Beiyi Zhu, Zhenbo Song et al.

Generating multiview-consistent $360^\circ$ ground-level scenes from satellite imagery is a challenging task with broad applications in simulation, autonomous navigation, and digital twin cities. Existing approaches primarily focus on synthesizing individual ground-view panoramas, often relying on auxiliary inputs like height maps or handcrafted projections, and struggle to produce multiview consistent sequences. In this paper, we propose SatDreamer360, a framework that generates geometrically consistent multi-view ground-level panoramas from a single satellite image, given a predefined pose trajectory. To address the large viewpoint discrepancy between ground and satellite images, we adopt a triplane representation to encode scene features and design a ray-based pixel attention mechanism that retrieves view-specific features from the triplane. To maintain multi-frame consistency, we introduce a panoramic epipolar-constrained attention module that aligns features across frames based on known relative poses. To support the evaluation, we introduce {VIGOR++}, a large-scale dataset for generating multi-view ground panoramas from a satellite image, by augmenting the original VIGOR dataset with more ground-view images and their pose annotations. Experiments show that SatDreamer360 outperforms existing methods in both satellite-to-ground alignment and multiview consistency.

Zhenbo Song, Wenhao Gao, Zhenyuan Zhang et al.

Deep learning-based face restoration models, increasingly prevalent in smart devices, have become targets for sophisticated backdoor attacks. These attacks, through subtle trigger injection into input face images, can lead to unexpected restoration outcomes. Unlike conventional methods focused on classification tasks, our approach introduces a unique degradation objective tailored for attacking restoration models. Moreover, we propose the Adaptive Selective Frequency Injection Backdoor Attack (AS-FIBA) framework, employing a neural network for input-specific trigger generation in the frequency domain, seamlessly blending triggers with benign images. This results in imperceptible yet effective attacks, guiding restoration predictions towards subtly degraded outputs rather than conspicuous targets. Extensive experiments demonstrate the efficacy of the degradation objective on state-of-the-art face restoration models. Additionally, it is notable that AS-FIBA can insert effective backdoors that are more imperceptible than existing backdoor attack methods, including WaNet, ISSBA, and FIBA.

Xinpeng Chen, Xiaofeng Han, Kaihao Zhang et al.

Layout design is a crucial step in developing mobile app pages. However, crafting satisfactory designs is time-intensive for designers: they need to consider which controls and content to present on the page, and then repeatedly adjust their size, position, and style for better aesthetics and structure. Although many design software can now help to perform these repetitive tasks, extensive training is needed to use them effectively. Moreover, collaborative design across app pages demands extra time to align standards and ensure consistent styling. In this work, we propose APD-agents, a large language model (LLM) driven multi-agent framework for automated page design in mobile applications. Our framework contains OrchestratorAgent, SemanticParserAgent, PrimaryLayoutAgent, TemplateRetrievalAgent, and RecursiveComponentAgent. Upon receiving the user's description of the page, the OrchestratorAgent can dynamically can direct other agents to accomplish users' design task. To be specific, the SemanticParserAgent is responsible for converting users' descriptions of page content into structured data. The PrimaryLayoutAgent can generate an initial coarse-grained layout of this page. The TemplateRetrievalAgent can fetch semantically relevant few-shot examples and enhance the quality of layout generation. Besides, a RecursiveComponentAgent can be used to decide how to recursively generate all the fine-grained sub-elements it contains for each element in the layout. Our work fully leverages the automatic collaboration capabilities of large-model-driven multi-agent systems. Experimental results on the RICO dataset show that our APD-agents achieve state-of-the-art performance.

Beibei Zhou, Zhiyuan Zhang, Zhenbo Song et al.

Robust and efficient deep LiDAR odometry models are crucial for accurate localization and 3D reconstruction, but typically require extensive and diverse training data to adapt to diverse environments, leading to inefficiencies. To tackle this, we introduce an active training framework designed to selectively extract training data from diverse environments, thereby reducing the training load and enhancing model generalization. Our framework is based on two key strategies: Initial Training Set Selection (ITSS) and Active Incremental Selection (AIS). ITSS begins by breaking down motion sequences from general weather into nodes and edges for detailed trajectory analysis, prioritizing diverse sequences to form a rich initial training dataset for training the base model. For complex sequences that are difficult to analyze, especially under challenging snowy weather conditions, AIS uses scene reconstruction and prediction inconsistency to iteratively select training samples, refining the model to handle a wide range of real-world scenarios. Experiments across datasets and weather conditions validate our approach's effectiveness. Notably, our method matches the performance of full-dataset training with just 52\% of the sequence volume, demonstrating the training efficiency and robustness of our active training paradigm. By optimizing the training process, our approach sets the stage for more agile and reliable LiDAR odometry systems, capable of navigating diverse environmental conditions with greater precision.

Zhenbo Song, Zhenyuan Zhang, Kaihao Zhang et al.

This study delves into the enhancement of Under-Display Camera (UDC) image restoration models, focusing on their robustness against adversarial attacks. Despite its innovative approach to seamless display integration, UDC technology faces unique image degradation challenges exacerbated by the susceptibility to adversarial perturbations. Our research initially conducts an in-depth robustness evaluation of deep-learning-based UDC image restoration models by employing several white-box and black-box attacking methods. This evaluation is pivotal in understanding the vulnerabilities of current UDC image restoration techniques. Following the assessment, we introduce a defense framework integrating adversarial purification with subsequent fine-tuning processes. First, our approach employs diffusion-based adversarial purification, effectively neutralizing adversarial perturbations. Then, we apply the fine-tuning methodologies to refine the image restoration models further, ensuring that the quality and fidelity of the restored images are maintained. The effectiveness of our proposed approach is validated through extensive experiments, showing marked improvements in resilience against typical adversarial attacks.

Zhaoxin Fan, Zhenbo Song, Wenping Zhang et al.

Point cloud-based large scale place recognition is an important but challenging task for many applications such as Simultaneous Localization and Mapping (SLAM). Taking the task as a point cloud retrieval problem, previous methods have made delightful achievements. However, how to deal with catastrophic collapse caused by rotation problems is still under-explored. In this paper, to tackle the issue, we propose a novel Point Cloud-based Rotation-aware Large Scale Place Recognition Network (RPR-Net). In particular, to solve the problem, we propose to learn rotation-invariant features in three steps. First, we design three kinds of novel Rotation-Invariant Features (RIFs), which are low-level features that can hold the rotation-invariant property. Second, using these RIFs, we design an attentive module to learn rotation-invariant kernels. Third, we apply these kernels to previous point cloud features to generate new features, which is the well-known SO(3) mapping process. By doing so, high-level scene-specific rotation-invariant features can be learned. We call the above process an Attentive Rotation-Invariant Convolution (ARIConv). To achieve the place recognition goal, we build RPR-Net, which takes ARIConv as a basic unit to construct a dense network architecture. Then, powerful global descriptors used for retrieval-based place recognition can be sufficiently extracted from RPR-Net. Experimental results on prevalent datasets show that our method achieves comparable results to existing state-of-the-art place recognition models and significantly outperforms other rotation-invariant baseline models when solving rotation problems.

Zhaoxin Fan, Zhenbo Song, Hongyan Liu et al.

Point cloud-based large scale place recognition is fundamental for many applications like Simultaneous Localization and Mapping (SLAM). Although many models have been proposed and have achieved good performance by learning short-range local features, long-range contextual properties have often been neglected. Moreover, the model size has also become a bottleneck for their wide applications. To overcome these challenges, we propose a super light-weight network model termed SVT-Net for large scale place recognition. Specifically, on top of the highly efficient 3D Sparse Convolution (SP-Conv), an Atom-based Sparse Voxel Transformer (ASVT) and a Cluster-based Sparse Voxel Transformer (CSVT) are proposed to learn both short-range local features and long-range contextual features in this model. Consisting of ASVT and CSVT, SVT-Net can achieve state-of-the-art on benchmark datasets in terms of both accuracy and speed with a super-light model size (0.9M). Meanwhile, two simplified versions of SVT-Net are introduced, which also achieve state-of-the-art and further reduce the model size to 0.8M and 0.4M respectively.

Peng Wan, Zhenbo Song, Jianfeng Lu

In this paper, we present a novel end-to-end deep neural network model for autonomous driving that takes monocular image sequence as input, and directly generates the steering control angle. Firstly, we model the end-to-end driving problem as a local path planning process. Inspired by the environmental representation in the classical planning algorithms(i.e. the beam curvature method), pixel-wise orientations are fed into the network to learn direction-aware features. Next, to handle the imbalanced distribution of steering values in training datasets, we propose an improvement on a cost-sensitive loss function named SteeringLoss2. Besides, we also present a new end-to-end driving dataset, which provides corresponding LiDAR and image sequences, as well as standard driving behaviors. Our dataset includes multiple driving scenarios, such as urban, country, and off-road. Numerous experiments are conducted on both public available LiVi-Set and our own dataset, and the results show that the model using our proposed methods can predict steering angle accurately.

Zhenbo Song, Jianfeng Lu, Tong Zhang et al.

Inter-vehicle distance and relative velocity estimations are two basic functions for any ADAS (Advanced driver-assistance systems). In this paper, we propose a monocular camera-based inter-vehicle distance and relative velocity estimation method based on end-to-end training of a deep neural network. The key novelty of our method is the integration of multiple visual clues provided by any two time-consecutive monocular frames, which include deep feature clue, scene geometry clue, as well as temporal optical flow clue. We also propose a vehicle-centric sampling mechanism to alleviate the effect of perspective distortion in the motion field (i.e. optical flow). We implement the method by a light-weight deep neural network. Extensive experiments are conducted which confirm the superior performance of our method over other state-of-the-art methods, in terms of estimation accuracy, computational speed, and memory footprint.