Dong Chen

Abdelrahman Abouelenin, Atabak Ashfaq, Adam Atkinson et al. · microsoft-research

We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of 200K tokens to better support multilingual applications, as well as group query attention for more efficient long-sequence generation. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model. Its novel modality extension approach leverages LoRA adapters and modality-specific routers to allow multiple inference modes combining various modalities without interference. For example, it now ranks first in the OpenASR leaderboard to date, although the LoRA component of the speech/audio modality has just 460 million parameters. Phi-4-Multimodal supports scenarios involving (vision + language), (vision + speech), and (speech/audio) inputs, outperforming larger vision-language and speech-language models on a wide range of tasks. Additionally, we experiment to further train Phi-4-Mini to enhance its reasoning capabilities. Despite its compact 3.8-billion-parameter size, this experimental version achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.

Tengfei Wang, Bo Zhang, Ting Zhang et al. · microsoft-research

This paper presents a 3D generative model that uses diffusion models to automatically generate 3D digital avatars represented as neural radiance fields. A significant challenge in generating such avatars is that the memory and processing costs in 3D are prohibitive for producing the rich details required for high-quality avatars. To tackle this problem we propose the roll-out diffusion network (Rodin), which represents a neural radiance field as multiple 2D feature maps and rolls out these maps into a single 2D feature plane within which we perform 3D-aware diffusion. The Rodin model brings the much-needed computational efficiency while preserving the integrity of diffusion in 3D by using 3D-aware convolution that attends to projected features in the 2D feature plane according to their original relationship in 3D. We also use latent conditioning to orchestrate the feature generation for global coherence, leading to high-fidelity avatars and enabling their semantic editing based on text prompts. Finally, we use hierarchical synthesis to further enhance details. The 3D avatars generated by our model compare favorably with those produced by existing generative techniques. We can generate highly detailed avatars with realistic hairstyles and facial hair like beards. We also demonstrate 3D avatar generation from image or text as well as text-guided editability.

Xiaoyi Dong, Jianmin Bao, Yinglin Zheng et al.

This paper presents a simple yet effective framework MaskCLIP, which incorporates a newly proposed masked self-distillation into contrastive language-image pretraining. The core idea of masked self-distillation is to distill representation from a full image to the representation predicted from a masked image. Such incorporation enjoys two vital benefits. First, masked self-distillation targets local patch representation learning, which is complementary to vision-language contrastive focusing on text-related representation. Second, masked self-distillation is also consistent with vision-language contrastive from the perspective of training objective as both utilize the visual encoder for feature aligning, and thus is able to learn local semantics getting indirect supervision from the language. We provide specially designed experiments with a comprehensive analysis to validate the two benefits. Symmetrically, we also introduce the local semantic supervision into the text branch, which further improves the pretraining performance. With extensive experiments, we show that MaskCLIP, when applied to various challenging downstream tasks, achieves superior results in linear probing, finetuning, and zero-shot performance with the guidance of the language encoder. Code will be release at \url{https://github.com/LightDXY/MaskCLIP}.

Binxin Yang, Shuyang Gu, Bo Zhang et al. · microsoft-research

Language-guided image editing has achieved great success recently. In this paper, for the first time, we investigate exemplar-guided image editing for more precise control. We achieve this goal by leveraging self-supervised training to disentangle and re-organize the source image and the exemplar. However, the naive approach will cause obvious fusing artifacts. We carefully analyze it and propose an information bottleneck and strong augmentations to avoid the trivial solution of directly copying and pasting the exemplar image. Meanwhile, to ensure the controllability of the editing process, we design an arbitrary shape mask for the exemplar image and leverage the classifier-free guidance to increase the similarity to the exemplar image. The whole framework involves a single forward of the diffusion model without any iterative optimization. We demonstrate that our method achieves an impressive performance and enables controllable editing on in-the-wild images with high fidelity.

Tiankai Hang, Shuyang Gu, Chen Li et al.

Denoising diffusion models have been a mainstream approach for image generation, however, training these models often suffers from slow convergence. In this paper, we discovered that the slow convergence is partly due to conflicting optimization directions between timesteps. To address this issue, we treat the diffusion training as a multi-task learning problem, and introduce a simple yet effective approach referred to as Min-SNR-$γ$. This method adapts loss weights of timesteps based on clamped signal-to-noise ratios, which effectively balances the conflicts among timesteps. Our results demonstrate a significant improvement in converging speed, 3.4$\times$ faster than previous weighting strategies. It is also more effective, achieving a new record FID score of 2.06 on the ImageNet $256\times256$ benchmark using smaller architectures than that employed in previous state-of-the-art. The code is available at https://github.com/TiankaiHang/Min-SNR-Diffusion-Training.

Wengang Zhou, Weilun Wang, Jianmin Bao et al.

Denoising Diffusion Probabilistic Models (DDPMs) have achieved remarkable success in various image generation tasks compared with Generative Adversarial Nets (GANs). Recent work on semantic image synthesis mainly follows the de facto GAN-based approaches, which may lead to unsatisfactory quality or diversity of generated images. In this paper, we propose a novel framework based on DDPM for semantic image synthesis. Unlike previous conditional diffusion model directly feeds the semantic layout and noisy image as input to a U-Net structure, which may not fully leverage the information in the input semantic mask, our framework processes semantic layout and noisy image differently. It feeds noisy image to the encoder of the U-Net structure while the semantic layout to the decoder by multi-layer spatially-adaptive normalization operators. To further improve the generation quality and semantic interpretability in semantic image synthesis, we introduce the classifier-free guidance sampling strategy, which acknowledge the scores of an unconditional model for sampling process. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our proposed method, achieving state-of-the-art performance in terms of fidelity (FID) and diversity (LPIPS). Our code and pretrained models are available at https://github.com/WeilunWang/semantic-diffusion-model.

Xiaoyi Dong, Jianmin Bao, Dongdong Chen et al.

In this work we propose Identity Consistency Transformer, a novel face forgery detection method that focuses on high-level semantics, specifically identity information, and detecting a suspect face by finding identity inconsistency in inner and outer face regions. The Identity Consistency Transformer incorporates a consistency loss for identity consistency determination. We show that Identity Consistency Transformer exhibits superior generalization ability not only across different datasets but also across various types of image degradation forms found in real-world applications including deepfake videos. The Identity Consistency Transformer can be easily enhanced with additional identity information when such information is available, and for this reason it is especially well-suited for detecting face forgeries involving celebrities. Code will be released at \url{https://github.com/LightDXY/ICT_DeepFake}

Junshu Tang, Tengfei Wang, Bo Zhang et al. · microsoft-research

In this work, we investigate the problem of creating high-fidelity 3D content from only a single image. This is inherently challenging: it essentially involves estimating the underlying 3D geometry while simultaneously hallucinating unseen textures. To address this challenge, we leverage prior knowledge from a well-trained 2D diffusion model to act as 3D-aware supervision for 3D creation. Our approach, Make-It-3D, employs a two-stage optimization pipeline: the first stage optimizes a neural radiance field by incorporating constraints from the reference image at the frontal view and diffusion prior at novel views; the second stage transforms the coarse model into textured point clouds and further elevates the realism with diffusion prior while leveraging the high-quality textures from the reference image. Extensive experiments demonstrate that our method outperforms prior works by a large margin, resulting in faithful reconstructions and impressive visual quality. Our method presents the first attempt to achieve high-quality 3D creation from a single image for general objects and enables various applications such as text-to-3D creation and texture editing.

Tengfei Wang, Ting Zhang, Bo Zhang et al. · microsoft-research

We propose to use pretraining to boost general image-to-image translation. Prior image-to-image translation methods usually need dedicated architectural design and train individual translation models from scratch, struggling for high-quality generation of complex scenes, especially when paired training data are not abundant. In this paper, we regard each image-to-image translation problem as a downstream task and introduce a simple and generic framework that adapts a pretrained diffusion model to accommodate various kinds of image-to-image translation. We also propose adversarial training to enhance the texture synthesis in the diffusion model training, in conjunction with normalized guidance sampling to improve the generation quality. We present extensive empirical comparison across various tasks on challenging benchmarks such as ADE20K, COCO-Stuff, and DIODE, showing the proposed pretraining-based image-to-image translation (PITI) is capable of synthesizing images of unprecedented realism and faithfulness.

Xiaoyi Dong, Jianmin Bao, Ting Zhang et al.

Recent studies have shown that CLIP has achieved remarkable success in performing zero-shot inference while its fine-tuning performance is not satisfactory. In this paper, we identify that fine-tuning performance is significantly impacted by hyper-parameter choices. We examine various key hyper-parameters and empirically evaluate their impact in fine-tuning CLIP for classification tasks through a comprehensive study. We find that the fine-tuning performance of CLIP is substantially underestimated. Equipped with hyper-parameter refinement, we demonstrate CLIP itself is better or at least competitive in fine-tuning compared with large-scale supervised pre-training approaches or latest works that use CLIP as prediction targets in Masked Image Modeling. Specifically, CLIP ViT-Base/16 and CLIP ViT-Large/14 can achieve 85.7%,88.0% finetuning Top-1 accuracy on the ImageNet-1K dataset . These observations challenge the conventional conclusion that CLIP is not suitable for fine-tuning, and motivate us to rethink recently proposed improvements based on CLIP. We will release our code publicly at \url{https://github.com/LightDXY/FT-CLIP}.

Bowen Zhang, Chenyang Qi, Pan Zhang et al. · microsoft-research

In this work, we propose an ID-preserving talking head generation framework, which advances previous methods in two aspects. First, as opposed to interpolating from sparse flow, we claim that dense landmarks are crucial to achieving accurate geometry-aware flow fields. Second, inspired by face-swapping methods, we adaptively fuse the source identity during synthesis, so that the network better preserves the key characteristics of the image portrait. Although the proposed model surpasses prior generation fidelity on established benchmarks, to further make the talking head generation qualified for real usage, personalized fine-tuning is usually needed. However, this process is rather computationally demanding that is unaffordable to standard users. To solve this, we propose a fast adaptation model using a meta-learning approach. The learned model can be adapted to a high-quality personalized model as fast as 30 seconds. Last but not the least, a spatial-temporal enhancement module is proposed to improve the fine details while ensuring temporal coherency. Extensive experiments prove the significant superiority of our approach over the state of the arts in both one-shot and personalized settings.

Gwangbin Bae, Martin de La Gorce, Tadas Baltrusaitis et al. · cambridge

State-of-the-art face recognition models show impressive accuracy, achieving over 99.8% on Labeled Faces in the Wild (LFW) dataset. Such models are trained on large-scale datasets that contain millions of real human face images collected from the internet. Web-crawled face images are severely biased (in terms of race, lighting, make-up, etc) and often contain label noise. More importantly, the face images are collected without explicit consent, raising ethical concerns. To avoid such problems, we introduce a large-scale synthetic dataset for face recognition, obtained by rendering digital faces using a computer graphics pipeline. We first demonstrate that aggressive data augmentation can significantly reduce the synthetic-to-real domain gap. Having full control over the rendering pipeline, we also study how each attribute (e.g., variation in facial pose, accessories and textures) affects the accuracy. Compared to SynFace, a recent method trained on GAN-generated synthetic faces, we reduce the error rate on LFW by 52.5% (accuracy from 91.93% to 96.17%). By fine-tuning the network on a smaller number of real face images that could reasonably be obtained with consent, we achieve accuracy that is comparable to the methods trained on millions of real face images.

Yidan Zhang, Ting Zhang, Dong Chen et al. · microsoft-research, pku

While generative modeling has become prevalent across numerous research fields, its integration into the realm of image retrieval remains largely unexplored and underjustified. In this paper, we present a novel methodology, reframing image retrieval as a variant of generative modeling and employing a sequence-to-sequence model. This approach is harmoniously aligned with the current trend towards unification in research, presenting a cohesive framework that allows for end-to-end differentiable searching. This, in turn, facilitates superior performance via direct optimization techniques. The development of our model, dubbed IRGen, addresses the critical technical challenge of converting an image into a concise sequence of semantic units, which is pivotal for enabling efficient and effective search. Extensive experiments demonstrate that our model achieves state-of-the-art performance on three widely-used image retrieval benchmarks as well as two million-scale datasets, yielding significant improvement compared to prior competitive retrieval methods. In addition, the notable surge in precision scores facilitated by generative modeling presents the potential to bypass the reranking phase, which is traditionally indispensable in practical retrieval workflows.

Junshu Tang, Bo Zhang, Binxin Yang et al. · microsoft-research

In contrast to the traditional avatar creation pipeline which is a costly process, contemporary generative approaches directly learn the data distribution from photographs. While plenty of works extend unconditional generative models and achieve some levels of controllability, it is still challenging to ensure multi-view consistency, especially in large poses. In this work, we propose a network that generates 3D-aware portraits while being controllable according to semantic parameters regarding pose, identity, expression and illumination. Our network uses neural scene representation to model 3D-aware portraits, whose generation is guided by a parametric face model that supports explicit control. While the latent disentanglement can be further enhanced by contrasting images with partially different attributes, there still exists noticeable inconsistency in non-face areas, e.g., hair and background, when animating expressions. Wesolve this by proposing a volume blending strategy in which we form a composite output by blending dynamic and static areas, with two parts segmented from the jointly learned semantic field. Our method outperforms prior arts in extensive experiments, producing realistic portraits with vivid expression in natural lighting when viewed from free viewpoints. It also demonstrates generalization ability to real images as well as out-of-domain data, showing great promise in real applications.

Daichi Horita, Jiaolong Yang, Dong Chen et al.

Image completion techniques have made significant progress in filling missing regions (i.e., holes) in images. However, large-hole completion remains challenging due to limited structural information. In this paper, we address this problem by integrating explicit structural guidance into diffusion-based image completion, forming our structure-guided diffusion model (SGDM). It consists of two cascaded diffusion probabilistic models: structure and texture generators. The structure generator generates an edge image representing plausible structures within the holes, which is then used for guiding the texture generation process. To train both generators jointly, we devise a novel strategy that leverages optimal Bayesian denoising, which denoises the output of the structure generator in a single step and thus allows backpropagation. Our diffusion-based approach enables a diversity of plausible completions, while the editable edges allow for editing parts of an image. Our experiments on natural scene (Places) and face (CelebA-HQ) datasets demonstrate that our method achieves a superior or comparable visual quality compared to state-of-the-art approaches. The code is available for research purposes at https://github.com/UdonDa/Structure_Guided_Diffusion_Model.

Hao Ouyang, Bo Zhang, Pan Zhang et al. · microsoft-research

We propose pose-guided multiplane image (MPI) synthesis which can render an animatable character in real scenes with photorealistic quality. We use a portable camera rig to capture the multi-view images along with the driving signal for the moving subject. Our method generalizes the image-to-image translation paradigm, which translates the human pose to a 3D scene representation -- MPIs that can be rendered in free viewpoints, using the multi-views captures as supervision. To fully cultivate the potential of MPI, we propose depth-adaptive MPI which can be learned using variable exposure images while being robust to inaccurate camera registration. Our method demonstrates advantageous novel-view synthesis quality over the state-of-the-art approaches for characters with challenging motions. Moreover, the proposed method is generalizable to novel combinations of training poses and can be explicitly controlled. Our method achieves such expressive and animatable character rendering all in real time, serving as a promising solution for practical applications.

Dong Chen, Lingfei Wu, Siliang Tang et al.

Recent years have seen a surge of interest in meta-learning techniques for tackling the few-shot learning (FSL) problem. However, the meta-learner is prone to overfitting since there are only a few available samples, which can be identified as sampling noise on a clean dataset. Moreover, when handling the data with noisy labels, the meta-learner could be extremely sensitive to label noise on a corrupted dataset. To address these two challenges, we present Eigen-Reptile (ER) that updates the meta-parameters with the main direction of historical task-specific parameters to alleviate sampling and label noise. Specifically, the main direction is computed in a fast way, where the scale of the calculated matrix is related to the number of gradient steps instead of the number of parameters. Furthermore, to obtain a more accurate main direction for Eigen-Reptile in the presence of many noisy labels, we further propose Introspective Self-paced Learning (ISPL). We have theoretically and experimentally demonstrated the soundness and effectiveness of the proposed Eigen-Reptile and ISPL. Particularly, our experiments on different tasks show that the proposed method is able to outperform or achieve highly competitive performance compared with other gradient-based methods with or without noisy labels. The code and data for the proposed method are provided for research purposes https://github.com/Anfeather/Eigen-Reptile.

Weilun Wang, Jianmin Bao, Wengang Zhou et al.

We present SinDiffusion, leveraging denoising diffusion models to capture internal distribution of patches from a single natural image. SinDiffusion significantly improves the quality and diversity of generated samples compared with existing GAN-based approaches. It is based on two core designs. First, SinDiffusion is trained with a single model at a single scale instead of multiple models with progressive growing of scales which serves as the default setting in prior work. This avoids the accumulation of errors, which cause characteristic artifacts in generated results. Second, we identify that a patch-level receptive field of the diffusion network is crucial and effective for capturing the image's patch statistics, therefore we redesign the network structure of the diffusion model. Coupling these two designs enables us to generate photorealistic and diverse images from a single image. Furthermore, SinDiffusion can be applied to various applications, i.e., text-guided image generation, and image outpainting, due to the inherent capability of diffusion models. Extensive experiments on a wide range of images demonstrate the superiority of our proposed method for modeling the patch distribution.

Zhicong Tang, Shuyang Gu, Jianmin Bao et al.

Vector quantized diffusion (VQ-Diffusion) is a powerful generative model for text-to-image synthesis, but sometimes can still generate low-quality samples or weakly correlated images with text input. We find these issues are mainly due to the flawed sampling strategy. In this paper, we propose two important techniques to further improve the sample quality of VQ-Diffusion. 1) We explore classifier-free guidance sampling for discrete denoising diffusion model and propose a more general and effective implementation of classifier-free guidance. 2) We present a high-quality inference strategy to alleviate the joint distribution issue in VQ-Diffusion. Finally, we conduct experiments on various datasets to validate their effectiveness and show that the improved VQ-Diffusion suppresses the vanilla version by large margins. We achieve an 8.44 FID score on MSCOCO, surpassing VQ-Diffusion by 5.42 FID score. When trained on ImageNet, we dramatically improve the FID score from 11.89 to 4.83, demonstrating the superiority of our proposed techniques.

Yiwei Ding, Wenjin Deng, Yinglin Zheng et al.

In this paper, we present the Intra- and Inter-Human Relation Networks (I^2R-Net) for Multi-Person Pose Estimation. It involves two basic modules. First, the Intra-Human Relation Module operates on a single person and aims to capture Intra-Human dependencies. Second, the Inter-Human Relation Module considers the relation between multiple instances and focuses on capturing Inter-Human interactions. The Inter-Human Relation Module can be designed very lightweight by reducing the resolution of feature map, yet learn useful relation information to significantly boost the performance of the Intra-Human Relation Module. Even without bells and whistles, our method can compete or outperform current competition winners. We conduct extensive experiments on COCO, CrowdPose, and OCHuman datasets. The results demonstrate that the proposed model surpasses all the state-of-the-art methods. Concretely, the proposed method achieves 77.4% AP on CrowPose dataset and 67.8% AP on OCHuman dataset respectively, outperforming existing methods by a large margin. Additionally, the ablation study and visualization analysis also prove the effectiveness of our model.

Hezhen Hu, Xiaoyi Dong, Jianmin Bao et al.

Pre-training is playing an increasingly important role in learning generic feature representation for Person Re-identification (ReID). We argue that a high-quality ReID representation should have three properties, namely, multi-level awareness, occlusion robustness, and cross-region invariance. To this end, we propose a simple yet effective pre-training framework, namely PersonMAE, which involves two core designs into masked autoencoders to better serve the task of Person Re-ID. 1) PersonMAE generates two regions from the given image with RegionA as the input and \textit{RegionB} as the prediction target. RegionA is corrupted with block-wise masking to mimic common occlusion in ReID and its remaining visible parts are fed into the encoder. 2) Then PersonMAE aims to predict the whole RegionB at both pixel level and semantic feature level. It encourages its pre-trained feature representations with the three properties mentioned above. These properties make PersonMAE compatible with downstream Person ReID tasks, leading to state-of-the-art performance on four downstream ReID tasks, i.e., supervised (holistic and occluded setting), and unsupervised (UDA and USL setting). Notably, on the commonly adopted supervised setting, PersonMAE with ViT-B backbone achieves 79.8% and 69.5% mAP on the MSMT17 and OccDuke datasets, surpassing the previous state-of-the-art by a large margin of +8.0 mAP, and +5.3 mAP, respectively.

Dong Chen, Duoqian Miao, Xuerong Zhao

In this paper, we point out that the essential differences between CNN-based and Transformer-based detectors, which cause the worse performance of small objects in Transformer-based methods, are the gap between local information and global dependencies in feature extraction and propagation. To address these differences, we propose a new vision Transformer, called Hybrid Network Transformer (Hyneter), after pre-experiments that indicate the gap causes CNN-based and Transformer-based methods to increase size-different objects result unevenly. Different from the divide and conquer strategy in previous methods, Hyneters consist of Hybrid Network Backbone (HNB) and Dual Switching module (DS), which integrate local information and global dependencies, and transfer them simultaneously. Based on the balance strategy, HNB extends the range of local information by embedding convolution layers into Transformer blocks, and DS adjusts excessive reliance on global dependencies outside the patch.

Yangyu Huang, Xi Chen, Jongyoo Kim et al.

Recent years have witnessed significant growth of face alignment. Though dense facial landmark is highly demanded in various scenarios, e.g., cosmetic medicine and facial beautification, most works only consider sparse face alignment. To address this problem, we present a framework that can enrich landmark density by existing sparse landmark datasets, e.g., 300W with 68 points and WFLW with 98 points. Firstly, we observe that the local patches along each semantic contour are highly similar in appearance. Then, we propose a weakly-supervised idea of learning the refinement ability on original sparse landmarks and adapting this ability to enriched dense landmarks. Meanwhile, several operators are devised and organized together to implement the idea. Finally, the trained model is applied as a plug-and-play module to the existing face alignment networks. To evaluate our method, we manually label the dense landmarks on 300W testset. Our method yields state-of-the-art accuracy not only in newly-constructed dense 300W testset but also in the original sparse 300W and WFLW testsets without additional cost.

Mengchen Liu, Qixiu Li, Dongdong Chen et al.

With the release of GPT-4V(O), its use in generating pseudo labels for multi-modality tasks has gained significant popularity. However, it is still a secret how to build such advanced models from its base large language models (LLMs). This work explores the potential of using LLMs alone for data generation and develop competitive multi-modality models focusing on chart understanding. We construct a large-scale chart dataset, SynChart, which contains approximately 4 million diverse chart images with over 75 million dense annotations, including data tables, code, descriptions, and question-answer sets. We trained a 4.2B chart-expert model using this dataset and achieve near-GPT-4O performance on the ChartQA task, surpassing GPT-4V.

Dengpan Fu, Dongdong Chen, Hao Yang et al.

This paper aims to address the problem of pre-training for person re-identification (Re-ID) with noisy labels. To setup the pre-training task, we apply a simple online multi-object tracking system on raw videos of an existing unlabeled Re-ID dataset "LUPerson" nd build the Noisy Labeled variant called "LUPerson-NL". Since theses ID labels automatically derived from tracklets inevitably contain noises, we develop a large-scale Pre-training framework utilizing Noisy Labels (PNL), which consists of three learning modules: supervised Re-ID learning, prototype-based contrastive learning, and label-guided contrastive learning. In principle, joint learning of these three modules not only clusters similar examples to one prototype, but also rectifies noisy labels based on the prototype assignment. We demonstrate that learning directly from raw videos is a promising alternative for pre-training, which utilizes spatial and temporal correlations as weak supervision. This simple pre-training task provides a scalable way to learn SOTA Re-ID representations from scratch on "LUPerson-NL" without bells and whistles. For example, by applying on the same supervised Re-ID method MGN, our pre-trained model improves the mAP over the unsupervised pre-training counterpart by 5.7%, 2.2%, 2.3% on CUHK03, DukeMTMC, and MSMT17 respectively. Under the small-scale or few-shot setting, the performance gain is even more significant, suggesting a better transferability of the learned representation. Code is available at https://github.com/DengpanFu/LUPerson-NL

Bowen Zhang, Shuyang Gu, Bo Zhang et al.

Despite the tantalizing success in a broad of vision tasks, transformers have not yet demonstrated on-par ability as ConvNets in high-resolution image generative modeling. In this paper, we seek to explore using pure transformers to build a generative adversarial network for high-resolution image synthesis. To this end, we believe that local attention is crucial to strike the balance between computational efficiency and modeling capacity. Hence, the proposed generator adopts Swin transformer in a style-based architecture. To achieve a larger receptive field, we propose double attention which simultaneously leverages the context of the local and the shifted windows, leading to improved generation quality. Moreover, we show that offering the knowledge of the absolute position that has been lost in window-based transformers greatly benefits the generation quality. The proposed StyleSwin is scalable to high resolutions, with both the coarse geometry and fine structures benefit from the strong expressivity of transformers. However, blocking artifacts occur during high-resolution synthesis because performing the local attention in a block-wise manner may break the spatial coherency. To solve this, we empirically investigate various solutions, among which we find that employing a wavelet discriminator to examine the spectral discrepancy effectively suppresses the artifacts. Extensive experiments show the superiority over prior transformer-based GANs, especially on high resolutions, e.g., 1024x1024. The StyleSwin, without complex training strategies, excels over StyleGAN on CelebA-HQ 1024, and achieves on-par performance on FFHQ-1024, proving the promise of using transformers for high-resolution image generation. The code and models will be available at https://github.com/microsoft/StyleSwin.

Yiting Cheng, Fangyun Wei, Jianmin Bao et al.

Domain adaptation for semantic segmentation enables to alleviate the need for large-scale pixel-wise annotations. Recently, self-supervised learning (SSL) with a combination of image-to-image translation shows great effectiveness in adaptive segmentation. The most common practice is to perform SSL along with image translation to well align a single domain (the source or target). However, in this single-domain paradigm, unavoidable visual inconsistency raised by image translation may affect subsequent learning. In this paper, based on the observation that domain adaptation frameworks performed in the source and target domain are almost complementary in terms of image translation and SSL, we propose a novel dual path learning (DPL) framework to alleviate visual inconsistency. Concretely, DPL contains two complementary and interactive single-domain adaptation pipelines aligned in source and target domain respectively. The inference of DPL is extremely simple, only one segmentation model in the target domain is employed. Novel technologies such as dual path image translation and dual path adaptive segmentation are proposed to make two paths promote each other in an interactive manner. Experiments on GTA5$\rightarrow$Cityscapes and SYNTHIA$\rightarrow$Cityscapes scenarios demonstrate the superiority of our DPL model over the state-of-the-art methods. The code and models are available at: \url{https://github.com/royee182/DPL}

Zhendong Wang, Jianmin Bao, Shuyang Gu et al.

In this paper, we present DesignDiffusion, a simple yet effective framework for the novel task of synthesizing design images from textual descriptions. A primary challenge lies in generating accurate and style-consistent textual and visual content. Existing works in a related task of visual text generation often focus on generating text within given specific regions, which limits the creativity of generation models, resulting in style or color inconsistencies between textual and visual elements if applied to design image generation. To address this issue, we propose an end-to-end, one-stage diffusion-based framework that avoids intricate components like position and layout modeling. Specifically, the proposed framework directly synthesizes textual and visual design elements from user prompts. It utilizes a distinctive character embedding derived from the visual text to enhance the input prompt, along with a character localization loss for enhanced supervision during text generation. Furthermore, we employ a self-play Direct Preference Optimization fine-tuning strategy to improve the quality and accuracy of the synthesized visual text. Extensive experiments demonstrate that DesignDiffusion achieves state-of-the-art performance in design image generation.

Bowen Zhang, Sicheng Xu, Chuxin Wang et al.

In this paper, we present a novel framework for video-to-4D generation that creates high-quality dynamic 3D content from single video inputs. Direct 4D diffusion modeling is extremely challenging due to costly data construction and the high-dimensional nature of jointly representing 3D shape, appearance, and motion. We address these challenges by introducing a Direct 4DMesh-to-GS Variation Field VAE that directly encodes canonical Gaussian Splats (GS) and their temporal variations from 3D animation data without per-instance fitting, and compresses high-dimensional animations into a compact latent space. Building upon this efficient representation, we train a Gaussian Variation Field diffusion model with temporal-aware Diffusion Transformer conditioned on input videos and canonical GS. Trained on carefully-curated animatable 3D objects from the Objaverse dataset, our model demonstrates superior generation quality compared to existing methods. It also exhibits remarkable generalization to in-the-wild video inputs despite being trained exclusively on synthetic data, paving the way for generating high-quality animated 3D content. Project page: https://gvfdiffusion.github.io/.

Jianzhu Huai, Yukai Lin, Yuan Zhuang et al.

Camera-IMU (Inertial Measurement Unit) sensor fusion has been extensively studied in recent decades. Numerous observability analysis and fusion schemes for motion estimation with self-calibration have been presented. However, it has been uncertain whether both camera and IMU intrinsic parameters are observable under general motion. To answer this question, by using the Lie derivatives, we first prove that for a rolling shutter (RS) camera-IMU system, all intrinsic and extrinsic parameters, camera time offset, and readout time of the RS camera, are observable with an unknown landmark. To our knowledge, we are the first to present such a proof. Next, to validate this analysis and to solve the drift issue of a structureless filter during standstills, we develop a Keyframe-based Sliding Window Filter (KSWF) for odometry and self-calibration, which works with a monocular RS camera or stereo RS cameras. Though the keyframe concept is widely used in vision-based sensor fusion, to our knowledge, KSWF is the first of its kind to support self-calibration. Our simulation and real data tests have validated that it is possible to fully calibrate the camera-IMU system using observations of opportunistic landmarks under diverse motion. Real data tests confirmed previous allusions that keeping landmarks in the state vector can remedy the drift in standstill, and showed that the keyframe-based scheme is an alternative solution.

Yinglin Zheng, Hao Yang, Ting Zhang et al.

How to learn a universal facial representation that boosts all face analysis tasks? This paper takes one step toward this goal. In this paper, we study the transfer performance of pre-trained models on face analysis tasks and introduce a framework, called FaRL, for general Facial Representation Learning in a visual-linguistic manner. On one hand, the framework involves a contrastive loss to learn high-level semantic meaning from image-text pairs. On the other hand, we propose exploring low-level information simultaneously to further enhance the face representation, by adding a masked image modeling. We perform pre-training on LAION-FACE, a dataset containing large amount of face image-text pairs, and evaluate the representation capability on multiple downstream tasks. We show that FaRL achieves better transfer performance compared with previous pre-trained models. We also verify its superiority in the low-data regime. More importantly, our model surpasses the state-of-the-art methods on face analysis tasks including face parsing and face alignment.

Shuyang Gu, Dong Chen, Jianmin Bao et al.

We present the vector quantized diffusion (VQ-Diffusion) model for text-to-image generation. This method is based on a vector quantized variational autoencoder (VQ-VAE) whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model (DDPM). We find that this latent-space method is well-suited for text-to-image generation tasks because it not only eliminates the unidirectional bias with existing methods but also allows us to incorporate a mask-and-replace diffusion strategy to avoid the accumulation of errors, which is a serious problem with existing methods. Our experiments show that the VQ-Diffusion produces significantly better text-to-image generation results when compared with conventional autoregressive (AR) models with similar numbers of parameters. Compared with previous GAN-based text-to-image methods, our VQ-Diffusion can handle more complex scenes and improve the synthesized image quality by a large margin. Finally, we show that the image generation computation in our method can be made highly efficient by reparameterization. With traditional AR methods, the text-to-image generation time increases linearly with the output image resolution and hence is quite time consuming even for normal size images. The VQ-Diffusion allows us to achieve a better trade-off between quality and speed. Our experiments indicate that the VQ-Diffusion model with the reparameterization is fifteen times faster than traditional AR methods while achieving a better image quality.

Pan Zhang, Bo Zhang, Ting Zhang et al.

Recent semi-supervised learning (SSL) methods are commonly based on pseudo labeling. Since the SSL performance is greatly influenced by the quality of pseudo labels, mutual learning has been proposed to effectively suppress the noises in the pseudo supervision. In this work, we propose robust mutual learning that improves the prior approach in two aspects. First, the vanilla mutual learners suffer from the coupling issue that models may converge to learn homogeneous knowledge. We resolve this issue by introducing mean teachers to generate mutual supervisions so that there is no direct interaction between the two students. We also show that strong data augmentations, model noises and heterogeneous network architectures are essential to alleviate the model coupling. Second, we notice that mutual learning fails to leverage the network's own ability for pseudo label refinement. Therefore, we introduce self-rectification that leverages the internal knowledge and explicitly rectifies the pseudo labels before the mutual teaching. Such self-rectification and mutual teaching collaboratively improve the pseudo label accuracy throughout the learning. The proposed robust mutual learning demonstrates state-of-the-art performance on semantic segmentation in low-data regime.

Yue Gao, Fangyun Wei, Jianmin Bao et al.

Cycle consistency is widely used for face editing. However, we observe that the generator tends to find a tricky way to hide information from the original image to satisfy the constraint of cycle consistency, making it impossible to maintain the rich details (e.g., wrinkles and moles) of non-editing areas. In this work, we propose a simple yet effective method named HifaFace to address the above-mentioned problem from two perspectives. First, we relieve the pressure of the generator to synthesize rich details by directly feeding the high-frequency information of the input image into the end of the generator. Second, we adopt an additional discriminator to encourage the generator to synthesize rich details. Specifically, we apply wavelet transformation to transform the image into multi-frequency domains, among which the high-frequency parts can be used to recover the rich details. We also notice that a fine-grained and wider-range control for the attribute is of great importance for face editing. To achieve this goal, we propose a novel attribute regression loss. Powered by the proposed framework, we achieve high-fidelity and arbitrary face editing, outperforming other state-of-the-art approaches.

Pan Zhang, Bo Zhang, Ting Zhang et al.

Self-training is a competitive approach in domain adaptive segmentation, which trains the network with the pseudo labels on the target domain. However inevitably, the pseudo labels are noisy and the target features are dispersed due to the discrepancy between source and target domains. In this paper, we rely on representative prototypes, the feature centroids of classes, to address the two issues for unsupervised domain adaptation. In particular, we take one step further and exploit the feature distances from prototypes that provide richer information than mere prototypes. Specifically, we use it to estimate the likelihood of pseudo labels to facilitate online correction in the course of training. Meanwhile, we align the prototypical assignments based on relative feature distances for two different views of the same target, producing a more compact target feature space. Moreover, we find that distilling the already learned knowledge to a self-supervised pretrained model further boosts the performance. Our method shows tremendous performance advantage over state-of-the-art methods. We will make the code publicly available.

Xiaoyi Dong, Jianmin Bao, Dongdong Chen et al.

DeepFake detection has so far been dominated by ``artifact-driven'' methods and the detection performance significantly degrades when either the type of image artifacts is unknown or the artifacts are simply too hard to find. In this work, we present an alternative approach: Identity-Driven DeepFake Detection. Our approach takes as input the suspect image/video as well as the target identity information (a reference image or video). We output a decision on whether the identity in the suspect image/video is the same as the target identity. Our motivation is to prevent the most common and harmful DeepFakes that spread false information of a targeted person. The identity-based approach is fundamentally different in that it does not attempt to detect image artifacts. Instead, it focuses on whether the identity in the suspect image/video is true. To facilitate research on identity-based detection, we present a new large scale dataset ``Vox-DeepFake", in which each suspect content is associated with multiple reference images collected from videos of a target identity. We also present a simple identity-based detection algorithm called the OuterFace, which may serve as a baseline for further research. Even trained without fake videos, the OuterFace algorithm achieves superior detection accuracy and generalizes well to different DeepFake methods, and is robust with respect to video degradation techniques -- a performance not achievable with existing detection algorithms.

Dengpan Fu, Dongdong Chen, Jianmin Bao et al.

In this paper, we present a large scale unlabeled person re-identification (Re-ID) dataset "LUPerson" and make the first attempt of performing unsupervised pre-training for improving the generalization ability of the learned person Re-ID feature representation. This is to address the problem that all existing person Re-ID datasets are all of limited scale due to the costly effort required for data annotation. Previous research tries to leverage models pre-trained on ImageNet to mitigate the shortage of person Re-ID data but suffers from the large domain gap between ImageNet and person Re-ID data. LUPerson is an unlabeled dataset of 4M images of over 200K identities, which is 30X larger than the largest existing Re-ID dataset. It also covers a much diverse range of capturing environments (eg, camera settings, scenes, etc.). Based on this dataset, we systematically study the key factors for learning Re-ID features from two perspectives: data augmentation and contrastive loss. Unsupervised pre-training performed on this large-scale dataset effectively leads to a generic Re-ID feature that can benefit all existing person Re-ID methods. Using our pre-trained model in some basic frameworks, our methods achieve state-of-the-art results without bells and whistles on four widely used Re-ID datasets: CUHK03, Market1501, DukeMTMC, and MSMT17. Our results also show that the performance improvement is more significant on small-scale target datasets or under few-shot setting.

Xingran Zhou, Bo Zhang, Ting Zhang et al.

We present the full-resolution correspondence learning for cross-domain images, which aids image translation. We adopt a hierarchical strategy that uses the correspondence from coarse level to guide the fine levels. At each hierarchy, the correspondence can be efficiently computed via PatchMatch that iteratively leverages the matchings from the neighborhood. Within each PatchMatch iteration, the ConvGRU module is employed to refine the current correspondence considering not only the matchings of larger context but also the historic estimates. The proposed CoCosNet v2, a GRU-assisted PatchMatch approach, is fully differentiable and highly efficient. When jointly trained with image translation, full-resolution semantic correspondence can be established in an unsupervised manner, which in turn facilitates the exemplar-based image translation. Experiments on diverse translation tasks show that CoCosNet v2 performs considerably better than state-of-the-art literature on producing high-resolution images.

Shuyang Gu, Jianmin Bao, Dong Chen

A key challenge in video enhancement and action recognition is to fuse useful information from neighboring frames. Recent works suggest establishing accurate correspondences between neighboring frames before fusing temporal information. However, the generated results heavily depend on the quality of correspondence estimation. In this paper, we propose a more robust solution: \emph{sampling and fusing multi-level features} across neighborhood frames to generate the results. Based on this idea, we introduce a new module to improve the capability of 3D convolution, namely, learnable sampling 3D convolution (\emph{LS3D-Conv}). We add learnable 2D offsets to 3D convolution which aims to sample locations on spatial feature maps across frames. The offsets can be learned for specific tasks. The \emph{LS3D-Conv} can flexibly replace 3D convolution layers in existing 3D networks and get new architectures, which learns the sampling at multiple feature levels. The experiments on video interpolation, video super-resolution, video denoising, and action recognition demonstrate the effectiveness of our approach.

Ziyu Wan, Bo Zhang, Dongdong Chen et al.

We propose to restore old photos that suffer from severe degradation through a deep learning approach. Unlike conventional restoration tasks that can be solved through supervised learning, the degradation in real photos is complex and the domain gap between synthetic images and real old photos makes the network fail to generalize. Therefore, we propose a novel triplet domain translation network by leveraging real photos along with massive synthetic image pairs. Specifically, we train two variational autoencoders (VAEs) to respectively transform old photos and clean photos into two latent spaces. And the translation between these two latent spaces is learned with synthetic paired data. This translation generalizes well to real photos because the domain gap is closed in the compact latent space. Besides, to address multiple degradations mixed in one old photo, we design a global branch with apartial nonlocal block targeting to the structured defects, such as scratches and dust spots, and a local branch targeting to the unstructured defects, such as noises and blurriness. Two branches are fused in the latent space, leading to improved capability to restore old photos from multiple defects. Furthermore, we apply another face refinement network to recover fine details of faces in the old photos, thus ultimately generating photos with enhanced perceptual quality. With comprehensive experiments, the proposed pipeline demonstrates superior performance over state-of-the-art methods as well as existing commercial tools in terms of visual quality for old photos restoration.

Di Hu, Xuhong Li, Lichao Mou et al.

Aerial scene recognition is a fundamental task in remote sensing and has recently received increased interest. While the visual information from overhead images with powerful models and efficient algorithms yields considerable performance on scene recognition, it still suffers from the variation of ground objects, lighting conditions etc. Inspired by the multi-channel perception theory in cognition science, in this paper, for improving the performance on the aerial scene recognition, we explore a novel audiovisual aerial scene recognition task using both images and sounds as input. Based on an observation that some specific sound events are more likely to be heard at a given geographic location, we propose to exploit the knowledge from the sound events to improve the performance on the aerial scene recognition. For this purpose, we have constructed a new dataset named AuDio Visual Aerial sceNe reCognition datasEt (ADVANCE). With the help of this dataset, we evaluate three proposed approaches for transferring the sound event knowledge to the aerial scene recognition task in a multimodal learning framework, and show the benefit of exploiting the audio information for the aerial scene recognition. The source code is publicly available for reproducibility purposes.

Ziyu Wan, Bo Zhang, Dongdong Chen et al.

We propose to restore old photos that suffer from severe degradation through a deep learning approach. Unlike conventional restoration tasks that can be solved through supervised learning, the degradation in real photos is complex and the domain gap between synthetic images and real old photos makes the network fail to generalize. Therefore, we propose a novel triplet domain translation network by leveraging real photos along with massive synthetic image pairs. Specifically, we train two variational autoencoders (VAEs) to respectively transform old photos and clean photos into two latent spaces. And the translation between these two latent spaces is learned with synthetic paired data. This translation generalizes well to real photos because the domain gap is closed in the compact latent space. Besides, to address multiple degradations mixed in one old photo, we design a global branch with a partial nonlocal block targeting to the structured defects, such as scratches and dust spots, and a local branch targeting to the unstructured defects, such as noises and blurriness. Two branches are fused in the latent space, leading to improved capability to restore old photos from multiple defects. The proposed method outperforms state-of-the-art methods in terms of visual quality for old photos restoration.

Pan Zhang, Bo Zhang, Dong Chen et al.

We present a general framework for exemplar-based image translation, which synthesizes a photo-realistic image from the input in a distinct domain (e.g., semantic segmentation mask, or edge map, or pose keypoints), given an exemplar image. The output has the style (e.g., color, texture) in consistency with the semantically corresponding objects in the exemplar. We propose to jointly learn the crossdomain correspondence and the image translation, where both tasks facilitate each other and thus can be learned with weak supervision. The images from distinct domains are first aligned to an intermediate domain where dense correspondence is established. Then, the network synthesizes images based on the appearance of semantically corresponding patches in the exemplar. We demonstrate the effectiveness of our approach in several image translation tasks. Our method is superior to state-of-the-art methods in terms of image quality significantly, with the image style faithful to the exemplar with semantic consistency. Moreover, we show the utility of our method for several applications

Lingzhi Li, Jianmin Bao, Ting Zhang et al.

In this paper we propose a novel image representation called face X-ray for detecting forgery in face images. The face X-ray of an input face image is a greyscale image that reveals whether the input image can be decomposed into the blending of two images from different sources. It does so by showing the blending boundary for a forged image and the absence of blending for a real image. We observe that most existing face manipulation methods share a common step: blending the altered face into an existing background image. For this reason, face X-ray provides an effective way for detecting forgery generated by most existing face manipulation algorithms. Face X-ray is general in the sense that it only assumes the existence of a blending step and does not rely on any knowledge of the artifacts associated with a specific face manipulation technique. Indeed, the algorithm for computing face X-ray can be trained without fake images generated by any of the state-of-the-art face manipulation methods. Extensive experiments show that face X-ray remains effective when applied to forgery generated by unseen face manipulation techniques, while most existing face forgery detection or deepfake detection algorithms experience a significant performance drop.

Lingzhi Li, Jianmin Bao, Hao Yang et al.

In this work, we propose a novel two-stage framework, called FaceShifter, for high fidelity and occlusion aware face swapping. Unlike many existing face swapping works that leverage only limited information from the target image when synthesizing the swapped face, our framework, in its first stage, generates the swapped face in high-fidelity by exploiting and integrating the target attributes thoroughly and adaptively. We propose a novel attributes encoder for extracting multi-level target face attributes, and a new generator with carefully designed Adaptive Attentional Denormalization (AAD) layers to adaptively integrate the identity and the attributes for face synthesis. To address the challenging facial occlusions, we append a second stage consisting of a novel Heuristic Error Acknowledging Refinement Network (HEAR-Net). It is trained to recover anomaly regions in a self-supervised way without any manual annotations. Extensive experiments on wild faces demonstrate that our face swapping results are not only considerably more perceptually appealing, but also better identity preserving in comparison to other state-of-the-art methods.

Jinpeng Lin, Hao Yang, Dong Chen et al.

Face parsing computes pixel-wise label maps for different semantic components (e.g., hair, mouth, eyes) from face images. Existing face parsing literature have illustrated significant advantages by focusing on individual regions of interest (RoIs) for faces and facial components. However, the traditional crop-and-resize focusing mechanism ignores all contextual area outside the RoIs, and thus is not suitable when the component area is unpredictable, e.g. hair. Inspired by the physiological vision system of human, we propose a novel RoI Tanh-warping operator that combines the central vision and the peripheral vision together. It addresses the dilemma between a limited sized RoI for focusing and an unpredictable area of surrounding context for peripheral information. To this end, we propose a novel hybrid convolutional neural network for face parsing. It uses hierarchical local based method for inner facial components and global methods for outer facial components. The whole framework is simple and principled, and can be trained end-to-end. To facilitate future research of face parsing, we also manually relabel the training data of the HELEN dataset and will make it public. Experiments on both HELEN and LFW-PL benchmarks demonstrate that our method surpasses state-of-the-art methods.

Shuyang Gu, Jianmin Bao, Hao Yang et al.

Portrait editing is a popular subject in photo manipulation. The Generative Adversarial Network (GAN) advances the generating of realistic faces and allows more face editing. In this paper, we argue about three issues in existing techniques: diversity, quality, and controllability for portrait synthesis and editing. To address these issues, we propose a novel end-to-end learning framework that leverages conditional GANs guided by provided face masks for generating faces. The framework learns feature embeddings for every face component (e.g., mouth, hair, eye), separately, contributing to better correspondences for image translation, and local face editing. With the mask, our network is available to many applications, like face synthesis driven by mask, face Swap+ (including hair in swapping), and local manipulation. It can also boost the performance of face parsing a bit as an option of data augmentation.

Wei Hu, Gusi Te, Ju He et al.

Face anti-spoofing plays a crucial role in protecting face recognition systems from various attacks. Previous model-based and deep learning approaches achieve satisfactory performance for 2D face spoofs, but remain limited for more advanced 3D attacks such as vivid masks. In this paper, we address 3D face anti-spoofing via the proposed Hypergraph Convolutional Neural Networks (HGCNN). Firstly, we construct a computation-efficient and posture-invariant face representation with only a few key points on hypergraphs. The hypergraph representation is then fed into the designed HGCNN with hypergraph convolution for feature extraction, while the depth auxiliary is also exploited for 3D mask anti-spoofing. Further, we build a 3D face attack database with color, depth and infrared light information to overcome the deficiency of 3D face anti-spoofing data. Experiments show that our method achieves the state-of-the-art performance over widely used 3D and 2D databases as well as the proposed one under various tests.

Jianmin Bao, Dong Chen, Fang Wen et al.

We propose a framework based on Generative Adversarial Networks to disentangle the identity and attributes of faces, such that we can conveniently recombine different identities and attributes for identity preserving face synthesis in open domains. Previous identity preserving face synthesis processes are largely confined to synthesizing faces with known identities that are already in the training dataset. To synthesize a face with identity outside the training dataset, our framework requires one input image of that subject to produce an identity vector, and any other input face image to extract an attribute vector capturing, e.g., pose, emotion, illumination, and even the background. We then recombine the identity vector and the attribute vector to synthesize a new face of the subject with the extracted attribute. Our proposed framework does not need to annotate the attributes of faces in any way. It is trained with an asymmetric loss function to better preserve the identity and stabilize the training process. It can also effectively leverage large amounts of unlabeled training face images to further improve the fidelity of the synthesized faces for subjects that are not presented in the labeled training face dataset. Our experiments demonstrate the efficacy of the proposed framework. We also present its usage in a much broader set of applications including face frontalization, face attribute morphing, and face adversarial example detection.

Jianmin Bao, Dong Chen, Fang Wen et al.

We present variational generative adversarial networks, a general learning framework that combines a variational auto-encoder with a generative adversarial network, for synthesizing images in fine-grained categories, such as faces of a specific person or objects in a category. Our approach models an image as a composition of label and latent attributes in a probabilistic model. By varying the fine-grained category label fed into the resulting generative model, we can generate images in a specific category with randomly drawn values on a latent attribute vector. Our approach has two novel aspects. First, we adopt a cross entropy loss for the discriminative and classifier network, but a mean discrepancy objective for the generative network. This kind of asymmetric loss function makes the GAN training more stable. Second, we adopt an encoder network to learn the relationship between the latent space and the real image space, and use pairwise feature matching to keep the structure of generated images. We experiment with natural images of faces, flowers, and birds, and demonstrate that the proposed models are capable of generating realistic and diverse samples with fine-grained category labels. We further show that our models can be applied to other tasks, such as image inpainting, super-resolution, and data augmentation for training better face recognition models.