Xianfang Zeng

Zibo Zhao, Wen Liu, Xin Chen et al. · deepmind, tencent-ai

We present a novel alignment-before-generation approach to tackle the challenging task of generating general 3D shapes based on 2D images or texts. Directly learning a conditional generative model from images or texts to 3D shapes is prone to producing inconsistent results with the conditions because 3D shapes have an additional dimension whose distribution significantly differs from that of 2D images and texts. To bridge the domain gap among the three modalities and facilitate multi-modal-conditioned 3D shape generation, we explore representing 3D shapes in a shape-image-text-aligned space. Our framework comprises two models: a Shape-Image-Text-Aligned Variational Auto-Encoder (SITA-VAE) and a conditional Aligned Shape Latent Diffusion Model (ASLDM). The former model encodes the 3D shapes into the shape latent space aligned to the image and text and reconstructs the fine-grained 3D neural fields corresponding to given shape embeddings via the transformer-based decoder. The latter model learns a probabilistic mapping function from the image or text space to the latent shape space. Our extensive experiments demonstrate that our proposed approach can generate higher-quality and more diverse 3D shapes that better semantically conform to the visual or textural conditional inputs, validating the effectiveness of the shape-image-text-aligned space for cross-modality 3D shape generation.

Shiyu Liu, Yucheng Han, Peng Xing et al. · tsinghua

In recent years, image editing models have witnessed remarkable and rapid development. The recent unveiling of cutting-edge multimodal models such as GPT-4o and Gemini2 Flash has introduced highly promising image editing capabilities. These models demonstrate an impressive aptitude for fulfilling a vast majority of user-driven editing requirements, marking a significant advancement in the field of image manipulation. However, there is still a large gap between the open-source algorithm with these closed-source models. Thus, in this paper, we aim to release a state-of-the-art image editing model, called Step1X-Edit, which can provide comparable performance against the closed-source models like GPT-4o and Gemini2 Flash. More specifically, we adopt the Multimodal LLM to process the reference image and the user's editing instruction. A latent embedding has been extracted and integrated with a diffusion image decoder to obtain the target image. To train the model, we build a data generation pipeline to produce a high-quality dataset. For evaluation, we develop the GEdit-Bench, a novel benchmark rooted in real-world user instructions. Experimental results on GEdit-Bench demonstrate that Step1X-Edit outperforms existing open-source baselines by a substantial margin and approaches the performance of leading proprietary models, thereby making significant contributions to the field of image editing.

Guoqing Ma, Haoyang Huang, Kun Yan et al.

We present Step-Video-T2V, a state-of-the-art text-to-video pre-trained model with 30B parameters and the ability to generate videos up to 204 frames in length. A deep compression Variational Autoencoder, Video-VAE, is designed for video generation tasks, achieving 16x16 spatial and 8x temporal compression ratios, while maintaining exceptional video reconstruction quality. User prompts are encoded using two bilingual text encoders to handle both English and Chinese. A DiT with 3D full attention is trained using Flow Matching and is employed to denoise input noise into latent frames. A video-based DPO approach, Video-DPO, is applied to reduce artifacts and improve the visual quality of the generated videos. We also detail our training strategies and share key observations and insights. Step-Video-T2V's performance is evaluated on a novel video generation benchmark, Step-Video-T2V-Eval, demonstrating its state-of-the-art text-to-video quality when compared with both open-source and commercial engines. Additionally, we discuss the limitations of current diffusion-based model paradigm and outline future directions for video foundation models. We make both Step-Video-T2V and Step-Video-T2V-Eval available at https://github.com/stepfun-ai/Step-Video-T2V. The online version can be accessed from https://yuewen.cn/videos as well. Our goal is to accelerate the innovation of video foundation models and empower video content creators.

Jingjing Chang, Yixiao Fang, Peng Xing et al.

Text-to-image (T2I) models have garnered significant attention for generating high-quality images aligned with text prompts. However, rapid T2I model advancements reveal limitations in early benchmarks, lacking comprehensive evaluations, for example, the evaluation on reasoning, text rendering and style. Notably, recent state-of-the-art models, with their rich knowledge modeling capabilities, show promising results on the image generation problems requiring strong reasoning ability, yet existing evaluation systems have not adequately addressed this frontier. To systematically address these gaps, we introduce OneIG-Bench, a meticulously designed comprehensive benchmark framework for fine-grained evaluation of T2I models across multiple dimensions, including prompt-image alignment, text rendering precision, reasoning-generated content, stylization, and diversity. By structuring the evaluation, this benchmark enables in-depth analysis of model performance, helping researchers and practitioners pinpoint strengths and bottlenecks in the full pipeline of image generation. Specifically, OneIG-Bench enables flexible evaluation by allowing users to focus on a particular evaluation subset. Instead of generating images for the entire set of prompts, users can generate images only for the prompts associated with the selected dimension and complete the corresponding evaluation accordingly. Our codebase and dataset are now publicly available to facilitate reproducible evaluation studies and cross-model comparisons within the T2I research community.

Haoyang Huang, Guoqing Ma, Nan Duan et al.

We present Step-Video-TI2V, a state-of-the-art text-driven image-to-video generation model with 30B parameters, capable of generating videos up to 102 frames based on both text and image inputs. We build Step-Video-TI2V-Eval as a new benchmark for the text-driven image-to-video task and compare Step-Video-TI2V with open-source and commercial TI2V engines using this dataset. Experimental results demonstrate the state-of-the-art performance of Step-Video-TI2V in the image-to-video generation task. Both Step-Video-TI2V and Step-Video-TI2V-Eval are available at https://github.com/stepfun-ai/Step-Video-TI2V.

Cailin Zhuang, Ailin Huang, Wei Cheng et al.

Story visualization aims to generate coherent image sequences that faithfully depict a narrative and align with character references. Despite progress in generative models, existing benchmarks are narrow in scope, often limited to short prompts, no character reference, or single-image cases, and fall short of real-world storytelling complexity. This hinders a nuanced understanding of model capabilities and limitations. We present ViStoryBench, a comprehensive benchmark designed to evaluate story visualization models across diverse narrative structures, visual styles, and character settings. The benchmark features richly annotated multi-shot scripts derived from curated stories spanning literature, film, and folklore. Large language models assist in story summarization and script generation, with all outputs verified by humans to ensure coherence and fidelity. Character references are carefully curated to maintain intra-story consistency across varying artistic styles. To enable thorough evaluation, ViStoryBench introduces a set of automated metrics that assess character consistency, style similarity, prompt adherence, aesthetic quality, and generation artifacts such as copy-paste behavior. These metrics are validated through human studies, and used to benchmark a broad range of open-source and commercial models. ViStoryBench offers a high-fidelity, multi-dimensional evaluation suite that facilitates systematic analysis and fosters future progress in visual storytelling.

Teng Hu, Jiangning Zhang, Ran Yi et al.

Employing LLMs for visual generation has recently become a research focus. However, the existing methods primarily transfer the LLM architecture to visual generation but rarely investigate the fundamental differences between language and vision. This oversight may lead to suboptimal utilization of visual generation capabilities within the LLM framework. In this paper, we explore the characteristics of visual embedding space under the LLM framework and discover that the correlation between visual embeddings can help achieve more stable and robust generation results. We present IAR, an Improved AutoRegressive Visual Generation Method that enhances the training efficiency and generation quality of LLM-based visual generation models. Firstly, we propose a Codebook Rearrangement strategy that uses balanced k-means clustering algorithm to rearrange the visual codebook into clusters, ensuring high similarity among visual features within each cluster. Leveraging the rearranged codebook, we propose a Cluster-oriented Cross-entropy Loss that guides the model to correctly predict the cluster where the token is located. This approach ensures that even if the model predicts the wrong token index, there is a high probability the predicted token is located in the correct cluster, which significantly enhances the generation quality and robustness. Extensive experiments demonstrate that our method consistently enhances the model training efficiency and performance from 100M to 1.4B, reducing the training time by half while achieving the same FID. Additionally, our approach can be applied to various LLM-based visual generation models and adheres to the scaling law, providing a promising direction for future research in LLM-based visual generation. The code is available at: https://github.com/sjtuplayer/IAR.

Fukun Yin, Shiyu Liu, Yucheng Han et al.

Recent advances in image editing models have shown remarkable progress. A common architectural design couples a multimodal large language model (MLLM) encoder with a diffusion decoder, as seen in systems such as Step1X-Edit and Qwen-Image-Edit, where the MLLM encodes both the reference image and the instruction but remains frozen during training. In this work, we demonstrate that unlocking the reasoning capabilities of MLLM can further push the boundaries of editing models. Specifically, we explore two reasoning mechanisms, thinking and reflection, which enhance instruction understanding and editing accuracy. Based on that, our proposed framework enables image editing in a thinking-editing-reflection loop: the thinking mechanism leverages the world knowledge of MLLM to interpret abstract instructions, while the reflection reviews editing results, automatically corrects unintended manipulations, and identifies the stopping round. Extensive experiments demonstrate that our reasoning approach achieves significant performance gains, with improvements of ImgEdit (+4.3%), GEdit (+4.7%), and Kris (+8.2%) when initializing our DiT from the Step1X-Edit (ReasonEdit-S), and also outperforms previous open-source methods on both GEdit and Kris when integrated with Qwen-Image-Edit (ReasonEdit-Q).

Jiangning Zhang, Xianfang Zeng, Chao Xu et al.

Audio-guided face reenactment aims to generate a photorealistic face that has matched facial expression with the input audio. However, current methods can only reenact a special person once the model is trained or need extra operations such as 3D rendering and image post-fusion on the premise of generating vivid faces. To solve the above challenge, we propose a novel \emph{R}eal-time \emph{A}udio-guided \emph{M}ulti-face reenactment approach named \emph{APB2FaceV2}, which can reenact different target faces among multiple persons with corresponding reference face and drive audio signal as inputs. Enabling the model to be trained end-to-end and have a faster speed, we design a novel module named Adaptive Convolution (AdaConv) to infuse audio information into the network, as well as adopt a lightweight network as our backbone so that the network can run in real time on CPU and GPU. Comparison experiments prove the superiority of our approach than existing state-of-the-art methods, and further experiments demonstrate that our method is efficient and flexible for practical applications https://github.com/zhangzjn/APB2FaceV2

Xin Kong, Xuemeng Yang, Guangyao Zhai et al.

Due to the difficulty in generating the effective descriptors which are robust to occlusion and viewpoint changes, place recognition for 3D point cloud remains an open issue. Unlike most of the existing methods that focus on extracting local, global, and statistical features of raw point clouds, our method aims at the semantic level that can be superior in terms of robustness to environmental changes. Inspired by the perspective of humans, who recognize scenes through identifying semantic objects and capturing their relations, this paper presents a novel semantic graph based approach for place recognition. First, we propose a novel semantic graph representation for the point cloud scenes by reserving the semantic and topological information of the raw point cloud. Thus, place recognition is modeled as a graph matching problem. Then we design a fast and effective graph similarity network to compute the similarity. Exhaustive evaluations on the KITTI dataset show that our approach is robust to the occlusion as well as viewpoint changes and outperforms the state-of-the-art methods with a large margin. Our code is available at: \url{https://github.com/kxhit/SG_PR}.

Xianfang Zeng, Xin Chen, Zhongqi Qi et al.

This paper presents Paint3D, a novel coarse-to-fine generative framework that is capable of producing high-resolution, lighting-less, and diverse 2K UV texture maps for untextured 3D meshes conditioned on text or image inputs. The key challenge addressed is generating high-quality textures without embedded illumination information, which allows the textures to be re-lighted or re-edited within modern graphics pipelines. To achieve this, our method first leverages a pre-trained depth-aware 2D diffusion model to generate view-conditional images and perform multi-view texture fusion, producing an initial coarse texture map. However, as 2D models cannot fully represent 3D shapes and disable lighting effects, the coarse texture map exhibits incomplete areas and illumination artifacts. To resolve this, we train separate UV Inpainting and UVHD diffusion models specialized for the shape-aware refinement of incomplete areas and the removal of illumination artifacts. Through this coarse-to-fine process, Paint3D can produce high-quality 2K UV textures that maintain semantic consistency while being lighting-less, significantly advancing the state-of-the-art in texturing 3D objects.

Wei Cheng, Juncheng Mu, Xianfang Zeng et al.

Texturing is a crucial step in the 3D asset production workflow, which enhances the visual appeal and diversity of 3D assets. Despite recent advancements in Text-to-Texture (T2T) generation, existing methods often yield subpar results, primarily due to local discontinuities, inconsistencies across multiple views, and their heavy dependence on UV unwrapping outcomes. To tackle these challenges, we propose a novel generation-refinement 3D texturing framework called MVPaint, which can generate high-resolution, seamless textures while emphasizing multi-view consistency. MVPaint mainly consists of three key modules. 1) Synchronized Multi-view Generation (SMG). Given a 3D mesh model, MVPaint first simultaneously generates multi-view images by employing an SMG model, which leads to coarse texturing results with unpainted parts due to missing observations. 2) Spatial-aware 3D Inpainting (S3I). To ensure complete 3D texturing, we introduce the S3I method, specifically designed to effectively texture previously unobserved areas. 3) UV Refinement (UVR). Furthermore, MVPaint employs a UVR module to improve the texture quality in the UV space, which first performs a UV-space Super-Resolution, followed by a Spatial-aware Seam-Smoothing algorithm for revising spatial texturing discontinuities caused by UV unwrapping. Moreover, we establish two T2T evaluation benchmarks: the Objaverse T2T benchmark and the GSO T2T benchmark, based on selected high-quality 3D meshes from the Objaverse dataset and the entire GSO dataset, respectively. Extensive experimental results demonstrate that MVPaint surpasses existing state-of-the-art methods. Notably, MVPaint could generate high-fidelity textures with minimal Janus issues and highly enhanced cross-view consistency.

Yongliang Wu, Zonghui Li, Xinting Hu et al.

Recent advances in multi-modal generative models have enabled significant progress in instruction-based image editing. However, while these models produce visually plausible outputs, their capacity for knowledge-based reasoning editing tasks remains under-explored. In this paper, we introduce KRIS-Bench (Knowledge-based Reasoning in Image-editing Systems Benchmark), a diagnostic benchmark designed to assess models through a cognitively informed lens. Drawing from educational theory, KRIS-Bench categorizes editing tasks across three foundational knowledge types: Factual, Conceptual, and Procedural. Based on this taxonomy, we design 22 representative tasks spanning 7 reasoning dimensions and release 1,267 high-quality annotated editing instances. To support fine-grained evaluation, we propose a comprehensive protocol that incorporates a novel Knowledge Plausibility metric, enhanced by knowledge hints and calibrated through human studies. Empirical results on 10 state-of-the-art models reveal significant gaps in reasoning performance, highlighting the need for knowledge-centric benchmarks to advance the development of intelligent image editing systems.

Chongjun Tu, Lin Zhang, Pengtao Chen et al.

Multimodal Large Language Models (MLLMs) have shown remarkable capabilities in video content understanding but still struggle with fine-grained motion comprehension. To comprehensively assess the motion understanding ability of existing MLLMs, we introduce FAVOR-Bench, comprising 1,776 videos with structured manual annotations of various motions. Our benchmark includes both close-ended and open-ended tasks. For close-ended evaluation, we carefully design 8,184 multiple-choice question-answer pairs spanning six distinct sub-tasks. For open-ended evaluation, we develop both a novel cost-efficient LLM-free and a GPT-assisted caption assessment method, where the former can enhance benchmarking interpretability and reproducibility. Comprehensive experiments with 21 state-of-the-art MLLMs reveal significant limitations in their ability to comprehend and describe detailed temporal dynamics in video motions. To alleviate this limitation, we further build FAVOR-Train, a dataset consisting of 17,152 videos with fine-grained motion annotations. The results of finetuning Qwen2.5-VL on FAVOR-Train yield consistent improvements on motion-related tasks of TVBench, MotionBench and our FAVOR-Bench. Comprehensive assessment results demonstrate that the proposed FAVOR-Bench and FAVOR-Train provide valuable tools to the community for developing more powerful video understanding models. Project page: \href{https://favor-bench.github.io/}{https://favor-bench.github.io/}.

Xinyao Liao, Xianfang Zeng, Liao Wang et al.

We propose MotionAgent, enabling fine-grained motion control for text-guided image-to-video generation. The key technique is the motion field agent that converts motion information in text prompts into explicit motion fields, providing flexible and precise motion guidance. Specifically, the agent extracts the object movement and camera motion described in the text and converts them into object trajectories and camera extrinsics, respectively. An analytical optical flow composition module integrates these motion representations in 3D space and projects them into a unified optical flow. An optical flow adapter takes the flow to control the base image-to-video diffusion model for generating fine-grained controlled videos. The significant improvement in the Video-Text Camera Motion metrics on VBench indicates that our method achieves precise control over camera motion. We construct a subset of VBench to evaluate the alignment of motion information in the text and the generated video, outperforming other advanced models on motion generation accuracy.

Lin Zhang, Xianfang Zeng, Kangcong Li et al.

We propose SC-Captioner, a reinforcement learning framework that enables the self-correcting capability of image caption models. Our crucial technique lies in the design of the reward function to incentivize accurate caption corrections. Specifically, the predicted and reference captions are decomposed into object, attribute, and relation sets using scene-graph parsing algorithms. We calculate the set difference between sets of initial and self-corrected captions to identify added and removed elements. These elements are matched against the reference sets to calculate correctness bonuses for accurate refinements and mistake punishments for wrong additions and removals, thereby forming the final reward. For image caption quality assessment, we propose a set of metrics refined from CAPTURE that alleviate its incomplete precision evaluation and inefficient relation matching problems. Furthermore, we collect a fine-grained annotated image caption dataset, RefinedCaps, consisting of 6.5K diverse images from COCO dataset. Experiments show that applying SC-Captioner on large visual-language models can generate better image captions across various scenarios, significantly outperforming the direct preference optimization training strategy.

Jiaxu Zhang, Xianfang Zeng, Xin Chen et al.

This paper presents DreamDance, a novel character art animation framework capable of producing stable, consistent character and scene motion conditioned on precise camera trajectories. To achieve this, we re-formulate the animation task as two inpainting-based steps: Camera-aware Scene Inpainting and Pose-aware Video Inpainting. The first step leverages a pre-trained image inpainting model to generate multi-view scene images from the reference art and optimizes a stable large-scale Gaussian field, which enables coarse background video rendering with camera trajectories. However, the rendered video is rough and only conveys scene motion. To resolve this, the second step trains a pose-aware video inpainting model that injects the dynamic character into the scene video while enhancing background quality. Specifically, this model is a DiT-based video generation model with a gating strategy that adaptively integrates the character's appearance and pose information into the base background video. Through extensive experiments, we demonstrate the effectiveness and generalizability of DreamDance, producing high-quality and consistent character animations with remarkable camera dynamics.

Jun Chen, Liang Liu, Yong Liu et al.

The quantized neural network (QNN) is an efficient approach for network compression and can be widely used in the implementation of FPGAs. This paper proposes a novel learning framework for n-bit QNNs, whose weights are constrained to the power of two. To solve the gradient vanishing problem, we propose a reconstructed gradient function for QNNs in back-propagation algorithm that can directly get the real gradient rather than estimating an approximate gradient of the expected loss. We also propose a novel QNN structure named n-BQ-NN, which uses shift operation to replace the multiply operation and is more suitable for the inference on FPGAs. Furthermore, we also design a shift vector processing element (SVPE) array to replace all 16-bit multiplications with SHIFT operations in convolution operation on FPGAs. We also carry out comparable experiments to evaluate our framework. The experimental results show that the quantized models of ResNet, DenseNet and AlexNet through our learning framework can achieve almost the same accuracies with the original full-precision models. Moreover, when using our learning framework to train our n-BQ-NN from scratch, it can achieve state-of-the-art results compared with typical low-precision QNNs. Experiments on Xilinx ZCU102 platform show that our n-BQ-NN with our SVPE can execute 2.9 times faster than with the vector processing element (VPE) in inference. As the SHIFT operation in our SVPE array will not consume Digital Signal Processings (DSPs) resources on FPGAs, the experiments have shown that the use of SVPE array also reduces average energy consumption to 68.7% of the VPE array with 16-bit.

Xianfang Zeng, Yusu Pan, Mengmeng Wang et al.

Recent works have shown how realistic talking face images can be obtained under the supervision of geometry guidance, e.g., facial landmark or boundary. To alleviate the demand for manual annotations, in this paper, we propose a novel self-supervised hybrid model (DAE-GAN) that learns how to reenact face naturally given large amounts of unlabeled videos. Our approach combines two deforming autoencoders with the latest advances in the conditional generation. On the one hand, we adopt the deforming autoencoder to disentangle identity and pose representations. A strong prior in talking face videos is that each frame can be encoded as two parts: one for video-specific identity and the other for various poses. Inspired by that, we utilize a multi-frame deforming autoencoder to learn a pose-invariant embedded face for each video. Meanwhile, a multi-scale deforming autoencoder is proposed to extract pose-related information for each frame. On the other hand, the conditional generator allows for enhancing fine details and overall reality. It leverages the disentangled features to generate photo-realistic and pose-alike face images. We evaluate our model on VoxCeleb1 and RaFD dataset. Experiment results demonstrate the superior quality of reenacted images and the flexibility of transferring facial movements between identities.

Jiangning Zhang, Xianfang Zeng, Mengmeng Wang et al.

This paper presents a novel multi-identity face reenactment framework, named FReeNet, to transfer facial expressions from an arbitrary source face to a target face with a shared model. The proposed FReeNet consists of two parts: Unified Landmark Converter (ULC) and Geometry-aware Generator (GAG). The ULC adopts an encode-decoder architecture to efficiently convert expression in a latent landmark space, which significantly narrows the gap of the face contour between source and target identities. The GAG leverages the converted landmark to reenact the photorealistic image with a reference image of the target person. Moreover, a new triplet perceptual loss is proposed to force the GAG module to learn appearance and geometry information simultaneously, which also enriches facial details of the reenacted images. Further experiments demonstrate the superiority of our approach for generating photorealistic and expression-alike faces, as well as the flexibility for transferring facial expressions between identities.