The limited research on 3D animal pose estimation is attributed to data scarcity and perspective ambiguities, despite its significant applications in various fields including biology, medicine, and animation. To resolve data scarcity, we put forward an unsupervised method for estimating 3D animal pose with only 2D pose available alone. To overcome perspective ambiguities, we propose canonical pose, camera, and view consistency losses to represent geometric consistency constraints for self-supervised learning. Specifically, the input 2D pose is fed into the pose generator network and camera network, and then regressed to the 3D canonical pose and camera rotation, respectively. In the training phase, the regressed 3D canonical pose is subjected to random re-projection to synthesize new 2D poses, which are also decomposed into 3D canonical pose and camera rotation to form geometric consistency constraints. Experimental results demonstrate that the proposed method achieves the best performance in unsupervised monocular 3D animal pose estimation. The corresponding code is available at: https://github.com/maicao2018/GeoSelfPose. © 2025 Elsevier Ltd

Automated face recognition systems are vulnerable against various attacks, such as adversarial attacks, digital manipulation and physical spoofs. As a special case of digital manipulation attacks, face morphing draws increasing concerns due to such attacks generalizing well across diverse face recognition systems. However, the threat of face morphing attacks is underestimated due to the following characteristics of state-of-the-art morphing methods. (i) Their generated face images have low visual quality with artifacts, (ii) they fail to guarantee high similarity with contributing subjects, and (iii) they do not explicitly consider countering face morphing detection methods when constructing morphing attacks. Based on the observation that facial geometry information is vital in face recognition, we present in this paper a geometry-aware generative model (GAGM), which can realize more threatening attacks against human experts, face recognition and morphing attack detection. GAGM synthesizes morphs with the drive of both facial geometry and texture based on dual invertible networks, resulting in visually realistic and highly deceptive morphed face images. To circumvent morphing-attack detection, GAGM implements a fine-grained adversarial attack strategy to mislead the detection methods. Visualization results demonstrate that GAGM, compared to existing techniques, is capable of generating visually faultless facial morphs. Meanwhile, extensive quantitative experiments show that GAGM can significantly increase the attack success rate against face recognition and deceive various morphing attack detection models. © 2025 Elsevier B.V.

Mesh generation plays a crucial role in 3D content creation, as mesh is widely used in various industrial applications. Recent works have achieved impressive results but still face several issues, such as unrealistic patterns or pits on surfaces, thin parts missing, and incomplete structures. Most of these problems stem from the choice of shape representation or the capabilities of the generative network. To alleviate these, we extend PoNQ, a Quadric Error Metrics (QEM)-based representation, and propose a novel model, QEMesh, for high-quality mesh generation. PoNQ divides the shape surface into tiny patches, each represented by a point with its normal and QEM matrix, which preserves fine local geometry information. In our QEMesh, we regard these elements as generable parameters and design a unique latent diffusion model containing a novel multi-decoder VAE for PoNQ parameters generation. Given the latent code generated by the diffusion model, three parameter decoders produce several PoNQ parameters within each voxel cell, and an occupancy decoder predicts which voxel cells containing parameters to form the final shape. Extensive evaluations demonstrate that our method generates results with watertight surfaces and is comparable to state-of-the-art methods in several main metrics. © 2025 IEEE.

Despite the promising results obtained by existing deepfake face detection methods for within-dataset detection, they often fail to generalize effectively to new datasets. We hypothesize that identity, a significant feature in facial recognition, is a key factor affecting deepfake detection models' cross-dataset performance. In the feature space learned by a real/fake classifier, facial features may cluster based on identity rather than their authenticity, which undermines the classifier's ability to distinguish between real and fake images. This paper introduces a novel training approach called Identity-Agnostic Learning (IAL) for deepfake face detection. IAL trains the detection model with identity-agnostic manner. It thus guides model to pay attention to the identity-irrelevant features. Experimental results demonstrate that our method effectively enhances the overall generalizability of deepfake face detection models. © 2025 IEEE.

Reconstructing 3D models from single-view images is a long-standing problem in computer vision. The latest advances for single-image 3D reconstruction extract a textual description from the input image and further utilize it to synthesize 3D models. However, existing methods focus on capturing a single key attribute of the image (e.g., object type, artistic style) and fail to consider the multi-perspective information required for accurate 3D reconstruction, such as object shape and material properties. Besides, the reliance on Neural Radiance Fields hinders their ability to reconstruct intricate surfaces and texture details. In this work, we propose MTFusion, which leverages both image data and textual descriptions for high-fidelity 3D reconstruction. Our approach consists of two stages. First, we adopt a novel multi-word textual inversion technique to extract a detailed text description capturing the image’s characteristics. Then, we use this description and the image to generate a 3D model with FlexiCubes. Additionally, MTFusion enhances FlexiCubes by employing a special decoder network for Signed Distance Functions, leading to faster training and finer surface representation. Extensive evaluations demonstrate that our MTFusion surpasses existing image-to-3D methods on a wide range of synthetic and real-world images. Furthermore, the ablation study proves the effectiveness of our network designs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Metal implants in patients cause severe streaking artifacts in computed tomography (CT) images, significantly compromising image quality. Deep learning methods have been successfully applied to metal artifact reduction (MAR) in CT, but often result in overly smooth images, failing to reconstruct complex details accurately. In this paper, we propose a multilevel dual-domain progressive enhancement network with embedded attention for MAR, termed MUPO-Net. Our approach constructs a Contrast Weight Mapping (CWM) module that generates a weighted heatmap, allocating weights to different regions based on the influence of metal artifacts, and an ASR-Net (Attention-Embedded Sinogram Restoration Network) that utilizes these weights to better remove artifacts in sinogram domain. Additionally, an Image Detail Enhancement Network (IDE-Net) is proposed to restore fine texture details in CT images through multi-scale feature fusion. Extensive experiments on both synthetic and clinical datasets demonstrate the superior effectiveness of MUPO-Net compared to the state-of-the-art MAR techniques. © 2025 IEEE.

The contrastive learning paradigm has been widely used for image-language pre-training and extended to videotext tuning. These approaches aim to maximize the similarity between positive sample pairs while minimizing that of negative ones through an alignment objective. Their performance is highly affected by the definition of positive and negative pairs which depends on the granularity of label classification. This effect is particularly apparent in video action recognition, where different fine-grained actions may belong to a shared coarse label. Therefore, indiscriminately treating a video sample and labels that are not identical at the fine-grained level but share the same coarse label as negative pairs leads to pushing the sample apart from the cluster of its basic coarse action. Such conflict can potentially prevent the model from pulling the sample and its target label closer. For a balanced understanding of coarse and fine-grained distinctions, we propose the Granularity- Aware Contrastive Learning (GACon) framework to improve contrastive learning for fine-grained action recognition. This is achieved through (i) a refined definition of sample-label relation and alignment objectives, and (ii) the exchange of coarse and finegrained information between two granularity-distinct experts. Experiments on four benchmarks of fine-grained action recognition show the superiority of our proposed GACon compared to existing approaches. © 2025 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

Face morphing attacks circumvent face recognition systems (FRSs) by creating a morphed image that contains multiple identities. However, existing face morphing attack methods either sacrifice image quality or compromise the identity preservation capability. Consequently, these attacks fail to bypass FRSs verification well while still managing to deceive human observers. These methods typically rely on global information from contributing images, ignoring the detailed information from effective facial regions. To address the above issues, we propose a novel morphing attack method to improve the quality of morphed images and better preserve the contributing identities. Our proposed method leverages the hierarchical generative network to capture both local detailed and global consistency information. Additionally, a mask-guided image blending module is dedicated to removing artifacts from areas outside the face to improve the image's visual quality. The proposed attack method is compared to state-of-the-art methods on three public datasets in terms of FRSs' vulnerability, attack detectability, and image quality. The results show our method's potential threat of deceiving FRSs while being capable of passing multiple morphing attack detection (MAD) scenarios. © 2024 IEEE.

Generating high-quality meshes with complex structures and realistic surfaces is the primary goal of 3D generative models. Existing methods typically employ sequence data or deformable tetrahedral grids for mesh generation. However, sequence-based methods have difficulty producing complex structures with many faces due to memory limits. The deformable tetrahedral grid-based method MeshDiffusion fails to recover realistic surfaces due to the inherent ambiguity in deformable grids. We propose the GenUDC framework to address these challenges by leveraging the Unsigned Dual Contouring (UDC) as the mesh representation. UDC discretizes a mesh in a regular grid and divides it into the face and vertex parts, recovering both complex structures and fine details. As a result, the one-to-one mapping between UDC and mesh resolves the ambiguity problem. In addition, GenUDC adopts a two-stage, coarse-to-fine generative process for 3D mesh generation. It first generates the face part as a rough shape and then the vertex part to craft a detailed shape. Extensive evaluations demonstrate the superiority of UDC as a mesh representation and the favorable performance of GenUDC in mesh generation. The code and trained models are available at https://github.com/TrepangCat/GenUDC. © 2024 ACM.

人群计数作为智能视频监控系统的重要任务之一,在公共安全及商业领域都有十分重要的应用价值,近年来已经成为机器视觉和人工智能领域的研究热点。其主要目标是针对人群场景的输入图像,估计出场景的总人数。人群计数技术可以自动、高效地辅助公共场所中人群监管,预防人群拥堵、踩踏等异常事件的发生。同时,该技术也可以应用到其他相关领域,如车辆计数,城市规划、生态资源调配。近年来,基于深度学习的人群计数技术取得了显著的发展和进步,然而,当前大部分研究工作都是以提升人群计数算法在人群互遮挡、多尺度、非均匀人群分布等视觉挑战下的计数准确率为目标,较少探讨人群计数算法的数据标注成本与计数准确率的权衡问题。目前的人群计数算法的数据标注方式是逐一详尽地标记出场景中全部人员,对于上千人场景的密集人群图片,这样的标注方式极其耗时、耗力。本文主要针对人群计数算法的高标注成本问题进行了深入分析,重点研究了利用低标注成本的点标注、区域人数标注及无标注数据有效实现人群计数。概括而言,本文的主要研究成果包括:（1）提出“点入框出”人群检测及计数算法。基于密度图回归的人群计数方法无法获得人员个体的尺度及位置输出,而在诸如人群追踪、仿真、异常行为预测等高层人群分析任务中人员个体的位置及尺度至关重要。目标检测方法可以获得人员个体的位置及尺度输出,然而,检测方法依赖于复杂、高成本的矩形框式标注。因此本文提出了使用简单、低成本的人头点式标注的“点入框出”人群检测及计数算法框架,该算法能够准确预测出人群场景中个体的尺度及位置并统计人群数目。该算法有效地利用人群场景先验知识,首先由人头点式弱标注生成人头矩形框式伪标注,再由伪标注在线更新,迭代选取较准确的矩形框伪标注,联合局部尺度约束回归损失函数及课程学习策略训练检测网络。多个人群计数数据集上的实验结果证明了本文提出算法的有效性。此外,在WIDER FACE人脸及TRANCOS车辆等目标计数数据集的实验结果证明了本文提出方法的通用性。（2）提出基于概率有序分类的人群计数算法。现有人群计数方法使用人头点或人头矩形框形式的实例级数据标注,实际部署时不高效。本文因此设计了基于概率有序分类的人群计数算法框架,该框架将子区域人数转化为密度等级类别学习分类网络。算法利用类别间潜在的序次知识,用序次先验约束结合隐表示概率建模,优化分类网络隐表示的学习;算法利用人数统计知识,提出可学习权重幅值分类器处理样本不平衡分布时的有偏学习问题;算法测试时将分类网络的预测转化为人数值实现人群计数。在Shanghai Tech,UCF-QNRF等通用人群计数数据集的实验结果证明了本文提出方法的有效性。（3）提出基于检测-回归双向知识迁移的跨域人群计数算法。现有的人群计数方法直接迁移到未见场合（目标域）数据上时性能大幅下降,为提升方法的跨域表现同时避免陷入高代价的数据标注困局,本文研究使用无标注目标域数据的跨域人群计数新方法。提出基于检测-回归双向知识迁移的跨域人群计数算法框架,该算法框架充分利用了检测及回归两类计数模型的互补性,通过在源域数据上建模检测及回归模型间知识的双向变换过程,在目标域数据上迁移检测及回归模型的互补知识的方式,提升它们在目标域的计数性能。本文首先分析了检测及回归两类模型输出结果的互补性质,然后详细探讨了源域上检测及回归模型输出结果间双向变换的求解方法,及目标域上实现检测及回归模型互补知识迁移的学习方法。多个数据集的跨域实验结果证明了本文提出方法的有效性。