Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks151 from public data
- Interpreting the Latent Space of GANs for Semantic Face Editing1,275
This work proposes a novel framework, called InterFaceGAN, for semantic face editing by interpreting the latent semantics learned by GANs, and finds that the latent code of well-trained generative models actually learns a disentangled representation after linear transformations.
- In-Domain GAN Inversion for Real Image Editing697
An in-domain GAN inversion approach, which not only faithfully reconstructs the input image but also ensures the inverted code to be semantically meaningful for editing, which achieves satisfying real image reconstruction and facilitates various image editing tasks, significantly outperforming start-of-the-arts.
- Closed-Form Factorization of Latent Semantics in GANs664
This work examines the internal representation learned by GANs to reveal the underlying variation factors in an unsupervised manner and proposes a closedform factorization algorithm for latent semantic discovery by directly decomposing the pre-trained weights.
- AnyDoor: Zero-shot Object-level Image Customization539
This work presents AnyDoor, a diffusion-based image generator with the power to teleport target objects to new scenes at user-specified locations with desired shapes, and proposes to borrow knowledge from video datasets, which can observe various forms of a single object, leading to stronger model generalizability and robustness.
- Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels536
It is argued that every pixel matters to the model training, even its prediction is ambiguous, and an effective pipeline is developed to make sufficient use of unlabeled data to separate reliable and unreliable pixels via the entropy of predictions.
- Notice of Removal: VideoFusion: Decomposed Diffusion Models for High-Quality Video Generation442
This work presents a decomposed diffusion process via resolving the per-frame noise into a base noise that is shared among all frames and a residual noise that varies along the time axis, which surpasses both GAN-based and diffusion-based alternatives in high-quality video generation.
- Composer: Creative and Controllable Image Synthesis with Composable Conditions404
This work offers a new generation paradigm that allows flexible control of the output image, such as spatial layout and palette, while maintaining the synthesis quality and model creativity, and serves as a general framework and facilitates a wide range of classical generative tasks without retraining.
- Image Processing Using Multi-Code GAN Prior347
A novel approach is proposed, called mGANprior, to incorporate the well-trained GANs as effective prior to a variety of image processing tasks, by employing multiple latent codes to generate multiple feature maps at some intermediate layer of the generator and composing them with adaptive channel importance to recover the input image.
- SpatialTracker: Tracking Any 2D Pixels in 3D Space220
This work proposes to estimate point trajectories in 3D space to mitigate the issues caused by image projection, and achieves state-of-the-art tracking performance both qualitatively and quantitatively, particularly in challenging scenarios such as out-of-plane rotation.
- Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis216
This work shows that highly-structured semantic hierarchy emerges in the deep generative representations from the state-of-the-art GANs like StyleGAN and BigGAN, trained for scene synthesis, and quantifies the causality between the layer-wise activations and the semantics occurring in the output image.
- 211
- 182
- 3D-aware Image Synthesis via Learning Structural and Textural Representations145
This work proposes a novel framework, termed as VolumeGAN, for high-fidelity 3D-aware image synthesis, through explicitly learning a structural representation and a textural representation in a Generative Adversarial Network.
- Advancing Open-source World Models142
LingBot-World is presented, an open-sourced world simulator stemming from video generation that maintains high fidelity and robust dynamics in a broad spectrum of environments, including realism, scientific contexts, cartoon styles, and beyond.
- 4K4D: Real-Time 4D View Synthesis at 4K Resolution136
4K4D, a 4D point cloud representation that supports hardware rasterization and network pre-computation to enable unprecedented rendering speed with a high rendering quality, is proposed and a novel hybrid appearance model is designed that significantly boosts the rendering quality while preserving efficiency.
- Generative Hierarchical Features from Synthesizing Images126
This work considers the pre-trained StyleGAN generator as a learned loss function and utilizes its layer-wise representation to train a novel hierarchical encoder, termed as Generative Hierarchical Feature (GH-Feat), which has strong transferability to both generative and discriminative tasks.
- Learning Temporally Consistent Video Depth from Video Diffusion Priors118
This work reformulates depth prediction into a conditional generation problem to provide contextual information within a clip and across clips to provide cross-clip context and proposes a consistent context-aware training and inference strategy for arbitrarily long videos to provide cross-clip context.
- Cones 2: Customizable Image Synthesis with Multiple Subjects114
This work finds that the text embedding regarding the subject token already serves as a simple yet effective representation that supports arbitrary combinations without any model tuning and proposes to employ layout, a very abstract and easy-to-obtain prior, as the spatial guidance for subject arrangement.
- SpatialTrackerV2: 3D Point Tracking Made Easy101
SpatialTrackerV2 decomposes world-space 3D motion into scene geometry, camera ego-motion, and pixel-wise object motion, with a fully differentiable and end-to-end architecture, allowing scalable training across a wide range of datasets.
- Ranni: Taming Text-to-Image Diffusion for Accurate Instruction Following95
This work introduces a semantic panel as the middleware in decoding texts to images, supporting the generator to better follow instructions, and enhances a pre-trained T2I generator regarding its textual controllability.
- 92
- 84
- DisCoScene: Spatially Disentangled Generative Radiance Fields for Controllable 3D-aware Scene Synthesis83
The proposed model spatially disentangles the whole scene into object-centric generative radiance fields by learning on only 2D images with the global-local discrimination, and obtains the generation fidelity and editing flexibility of individual objects while being able to efficiently compose objects and the background into a complete scene.
- Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset82
Ditto is introduced, a holistic framework that fuses the creative diversity of a leading image editor with an in-context video generator, overcoming the limited scope of existing models and resolves the prohibitive cost-quality trade-off.
- InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Prior75
This work proposes to guide the point initialization with an image-conditioned depth completion model, which learns to directly restore the depth map based on the observed image, and surpasses existing alternatives with sufficiently better fidelity and efficiency under various complex scenarios.
- Deep Generative Models on 3D Representations: A Survey72
A thorough review of the development of 3D generation, including 3D shape generation and 3D-aware image synthesis, from the perspectives of both algorithms and more importantly representations is made.
- A Recipe for Scaling up Text-to-Video Generation with Text-free Videos70
A novel text-to-video generation framework, termed TF-T2V, which can directly learn with text-free videos and validate the effectiveness and generalizability of the ideology on both native text-to-video generation and compositional video synthesis paradigms.
- One-Shot Generative Domain Adaptation70
This work aims to transfer a Generative Adversarial Network pre-trained on one image domain to another domain referred to as few as just one reference image, and proposes to constrain the diversity of the latent space through truncation in the training process, alleviating the optimization difficulty.
- Cross-Model Pseudo-Labeling for Semi-Supervised Action Recognition66
This work proposes a more effective pseudo-labeling scheme, called Cross-Model Pseudo-Labeling (CMPL), which introduces a lightweight auxiliary network in addition to the primary backbone, and asks them to predict pseudo-labels for each other, and observes that these two models tend to learn complementary representations from the same video clips.
- HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives54
HoloCine is a model that generates entire scenes holistically to ensure global consistency from the first shot to the last, marking a pivotal shift from clip synthesis towards automated filmmaking, making end-to-end cinematic creation a tangible future.
- MagicQuill: An Intelligent Interactive Image Editing System51
The MagicQuill system starts with a streamlined yet functionally robust interface, enabling users to articulate their ideas with just a few strokes, then monitored by a multimodal large language model (MLLM) to anticipate user intentions in real time, bypassing the need for prompt entry.
- MANF brakes TLR4 signaling by competitively binding S100A8 with S100A9 to regulate macrophage phenotypes in hepatic fibrosis51
A mechanism by which MANF targets S100A8/A9-TLR4 as a “brake” on the upstream of NF-κB pathway, which exerts an impact on macrophage differentiation and shed light on hepatic fibrosis treatment is revealed.
- Balancing Logit Variation for Long-Tailed Semantic Segmentation49
Although with an embarrassingly simple implementation, this work introduces category-wise variation into the network predictions in the training phase such that an instance is no longer projected to a feature point, but a small region instead.
- Prometheus: 3D-Aware Latent Diffusion Models for Feed-Forward Text-to-3D Scene Generation46
Prometheus, a 3D-aware latent diffusion model for text-to-3D generation at both object and scene levels in seconds is introduced and an RGB-D latent space is introduced into 3D Gaussian generation to disentangle appearance and geometry information, enabling efficient feed-forward generation of 3D Gaussians with better fidelity and geometry.
- LeviTor: 3D Trajectory Oriented Image-to-Video Synthesis45
This work proposes a pioneering method for 3D trajectory control in image-to-video synthesis by abstracting object masks into a few cluster points, accompanied by the depth information and the instance information, which are finally fed into a video diffusion model as the control signal.
- EnvGS: Modeling View-Dependent Appearance with Environment Gaussian45
This work introduces EnvGS, a novel approach that employs a set of Gaussian primitives as an explicit 3D representation for capturing reflections of environments that produces significantly more detailed reflections, achieving the best rendering quality in real-time novel view synthesis.
- 44
- Benchmarking Large Vision-Language Models via Directed Scene Graph for Comprehensive Image Captioning40
A detailed caption benchmark, termed as CompreCap, to evaluate the visual context from a directed scene graph view, to assess the generated detailed captions from LVLMs on multiple levels, including the object-level coverage, the accuracy of attribute descriptions, the score of key relationships, etc.
- 3D-Aware Indoor Scene Synthesis with Depth Priors39
This work proposes a dual-path generator, where one path is responsible for depth generation, whose intermediate features are injected into the other path as the condition for appearance rendering, which eases the 3D-aware synthesis with explicit geometry information.
- 39
- MaPa: Text-driven Photorealistic Material Painting for 3D Shapes36
This paper decomposes a shape into a set of segments and designs a segment-controlled diffusion model to synthesize 2D images that are aligned with mesh parts to generate segment-wise procedural material graphs as the appearance representation.
- GaussianPrediction: Dynamic 3D Gaussian Prediction for Motion Extrapolation and Free View Synthesis36
GaussianPrediction is a novel framework that empowers 3D Gaussian representations with dynamic scene modeling and future scenario synthesis in dynamic environments, and can forecast future states from any viewpoint, using video observations of dynamic scenes.
- Improving GAN Equilibrium by Raising Spatial Awareness35
This work aims to alleviate the issue of D dominating the competition in GANs, and proposes to align the spatial awareness of G with the attention map induced from D, effectively lessen the information gap between D and G.
- FlashFace: Human Image Personalization with High-fidelity Identity Preservation34
This work presents FlashFace, a practical tool with which users can easily personalize their own photos on the fly by providing one or a few reference face images and a text prompt, benefiting from two subtle designs.
- Scanning Only Once: An End-to-end Framework for Fast Temporal Grounding in Long Videos34
This work proposes an end-to-end framework for fast temporal grounding, which is able to model an hours-long video with one-time network execution, and significantly outperforms state-of-the-arts, and achieves 14.6× / 102.8× higher efficiency respectively.
- DepthLab: From Partial to Complete33
This work bridges this gap with DepthLab, a foundation depth inpainting model powered by image diffusion priors that demonstrates resilience to depth-deficient regions, and faithfully preserves scale consistency with the conditioned known depth when filling in missing values.
- Framer: Interactive Frame Interpolation33
This work proposes Framer for interactive frame interpolation, which targets producing smoothly transitioning frames between two images as per user creativity, and supports customizing the transition process by tailoring the trajectory of some selected keypoints.
- Ready-to-React: Online Reaction Policy for Two-Character Interaction Generation32
This paper proposes an online reaction policy, called Ready-to-React, to generate the next character pose based on past observed motions, and demonstrates that this approach can be controlled by sparse signals, making it well-suited for VR and other online interactive environments.
- MangaNinja: Line Art Colorization with Precise Reference Following31
This work incorporates two thoughtful designs to ensure precise character detail transcription, including a patch shuffling module to facilitate correspondence learning between the reference color image and the target line art, and a point-driven control scheme to enable fine-grained color matching.
- HeadArtist: Text-conditioned 3D Head Generation with Self Score Distillation31
Experimental results show that the HeadArtist approach produces high-quality 3D head sculptures with rich geometry and photo-realistic appearance, which significantly outperforms state-of-the-art methods.
- 31
- LinkGAN: Linking GAN Latents to Pixels for Controllable Image Synthesis30
An easy-to-use regularizer for GAN training, which helps explicitly link some axes of the latent space to a set of pixels in the synthesized image, which can improve the spatial controllability of both 2D and 3D-aware GAN models, barely sacrificing the synthesis performance.
- AniDoc: Animation Creation Made Easier29
AniDoc1 emerges as a video line art colorization tool, which automatically converts sketch sequences into colored animations following the reference character specification, yielding strong robustness to the variations between the reference character and each line art frame.
- 26
- 26
- 26
- Learning 3D-Aware Image Synthesis with Unknown Pose Distribution25
This work proposes PoF3D, a pose-free generator and the pose-aware discriminator that frees generative radiance fields from the requirements of 3D pose priors, and demonstrates the feasibility of learning high-quality 3D-aware image synthesis without using 3D posing priors for the first time.
- Rectified Diffusion Guidance for Conditional Generation22
The theory behind CFG is revisited and it is rigorously confirmed that the improper configuration of the combination coefficients brings about expectation shift of the generative distribution and a relaxation on the guidance coefficients is proposed such that denoising with ReCFG strictly aligns with the diffusion theory.
- 4D Gaussian Videos with Motion Layering21
This paper proposes a novel 4D Gaussian Video approach that enables the creation and streaming of photorealistic, volumetric videos for dynamic scenes over the Internet, and demonstrates that this method outperforms state-of-the-art volumetric video approaches with superior rendering quality and minimum storage overheads.
- 21
- 21
- 21
- Mimir: Improving Video Diffusion Models for Precise Text Understanding20
Mimir is an end-to-end training framework featuring a carefully tailored token fuser to harmonize the outputs from text encoders and LLMs, which allows the T2V model to fully leverage learned video priors while capitalizing on the text-related capability of LLMs.
- 20
- Exploring Sparse MoE in GANs for Text-conditioned Image Synthesis19
Aurora, a GAN-based text-to-image generator that employs a collection of experts to learn feature processing, together with a sparse router to adaptively select the most suitable expert for each feature point, closes the performance gap between GANs and industry-level diffusion models, maintaining a fast inference speed.
- Towards Smooth Video Composition19
This work investigates modeling the temporal relations for composing video with arbitrary length, from a few frames to even infinite, using generative adversarial networks (GANs), and develops a novel B-Spline based motion representation to ensure temporal smoothness to achieve infinite-length video generation.
- Towards More Accurate Diffusion Model Acceleration with a Timestep Tuner18
A timestep tuner that helps find a more accurate integral direction for a particular interval at the minimum cost and can be trained efficiently and boost the inference performance of various state-of-the-art acceleration methods, especially when there are few denoising steps.
- UKnow: A Unified Knowledge Protocol with Multimodal Knowledge Graph Datasets for Reasoning and Vision-Language Pre-Training18
This work categorizes data knowledge into five unit types, namely, in-image, in-text, cross-image, cross-text, and image-text, and sets up an efficient pipeline to help construct the multimodal knowledge graph from any data collection.
- Contextual AD Narration with Interleaved Multimodal Sequence17
This work proposes a simple and lightweight module that maps video features into the textual feature space and proposes a character-refinement module to provide more precise information by identifying the main characters who play more significant roles in the video context.
- Diffuman4D: 4D Consistent Human View Synthesis From Sparse-View Videos With Spatio-Temporal Diffusion Models17
This paper proposes a novel sliding iterative denoising process to enhance the spatio-temporal consistency of the 4D diffusion model, and is able to synthesize high-quality and consistent novel-view videos and significantly outperforms the existing approaches.
- NEAT: Distilling 3D Wireframes from Neural Attraction Fields17
The proposed NEAT enjoys the joint optimization of the neural fields and the global junctions from scratch, using view-dependent 2D observations without precomputed cross-view feature matching, demonstrating its superiority over state-of-the-art alternatives for 3D wireframe reconstruction.
- 17
- CompConv: A Compact Convolution Module for Efficient Feature Learning17
This work makes a close study of the convolution operator, which is the basic unit used in CNNs, to reduce its computing load and proposes a compact convolution module, called CompConv, to facilitate efficient feature learning.
- AvatarArtist: Open-Domain 4D Avatarization16
This work focuses on open-domain 4D avatarization, and selects parametric triplanes as the intermediate 4D representation, and proposes a practical training paradigm that takes advantage of both generative adversarial networks (GANs) and diffusion models.
- 16
- Using Unreliable Pseudo-Labels for Label-Efficient Semantic Segmentation15
This work separate reliable and unreliable pixels via the entropy of predictions, push each unreliable pixel to a category-wise queue that consists of negative keys, and manage to train the model with all candidate pixels through an effective pipeline to make sufficient use of unlabeled data.
- In-Domain GAN Inversion for Faithful Reconstruction and Editability14
In-domain GAN inversion is proposed, which consists of a domain-guided encoder and a domain-regularized optimizer, to regularize the inverted code in the native latent space of the pre-trained GAN model, to sufficiently reuse the knowledge learned by GANs for image reconstruction, facilitating a wide range of editing applications without any retraining.
- Regularized Mask Tuning: Uncovering Hidden Knowledge in Pre-trained Vision-Language Models14
This work designs a new type of tuning method, termed as regularized mask tuning, which masks the network parameters through a learnable selection, and is synergistic with most existing parameter-efficient tuning methods and can boost the performance on top of them.
- Dimensionality-Varying Diffusion Process14
This work makes a theoretical generalization of the forward diffusion process via signal decomposition that facilitates high-resolution image synthesis and improves FID of diffusion model trained on FFHQ at $1024\times 1024$ resolution from 52.40 to 10.46.
- 14
- Pulling Target to Source: A New Perspective on Domain Adaptive Semantic Segmentation13
This work proposes T2S-DA, which encourages the model to learn similar cross-domain features and shows that T2S-DA is quite qualified for the domain generalization task, verifying its domain-invariant property.
- Reprogramming tumor-associated macrophages and inhibiting tumor neovascularization by targeting MANF–HSF1–HSP70-1 pathway: An effective treatment for hepatocellular carcinoma12
This study analyzed differential genes in normal macrophages and TAMs and found that loss of MANF in TAMs accompanied by high levels of downstream genes negatively regulated by MANF promoted HCC progression in HCC patients and mice HCC model, especially tumor neovascularization.
- GLeaD: Improving GANs with A Generator-Leading Task12
This work proposes a new paradigm for adversarial training, which makes G assign a task to D as well, and expects D to extract representative features that can be adequately decoded by G to reconstruct the input.
- MANF Alleviates Sevoflurane-Induced Cognitive Impairment in Neonatal Mice by Modulating Microglial Activation and Polarization12
It is demonstrated that MANF ameliorated neurotoxicity induced by the sevoflurane anesthesia in young mice, and such protective effect was associated with inhibition of microglia activation and neuroinflammation.
- CCM: Adding Conditional Controls to Text-to-Image Consistency Models11
This technical report considers alternative strategies for adding ControlNet-like conditional control to CMs and presents three significant findings.
- Edicho: Consistent Image Editing in the Wild10
Edicho steps in with a training-free solution based on diffusion models, featuring a fundamental design principle of using explicit image correspondence to direct editing, and enjoys a plug-and-play nature and is compatible to most diffusion-based editing methods, such as ControlNet and BrushNet.
- Interspatial Attention for Efficient 4D Human Video Generation9
This paper introduces a new interspatial attention (ISA) mechanism as a scalable building block for modern diffusion transformer (DiT)-based video generation models and achieves state-of-the-art performance for 4D human video synthesis.
- MANF serves as a novel hepatocyte factor to promote liver regeneration after 2/3 partial hepatectomy via doubly targeting Wnt/β-catenin signaling9
This study confirms that MANF may serve as a novel hepatocyte factor that is closely linked to the activation of the Wnt/β-catenin pathway via intracellular and extracellular targets.
- SMaRt: Improving GANs with Score Matching Regularity9
This work revisit the mathematical foundations of GANs, and theoretically reveal that the native adversarial loss for GAN training is insufficient to fix the problem of subsets with positive Lebesgue measure of the generated data manifold lying out of the real data manifold, and finds that score matching serves as a promising solution to this issue.
- UIKA: Fast Universal Head Avatar from Pose-Free Images8
UIKA, a feed-forward animatable Gaussian head model from an arbitrary number of pose-free inputs, including a single image, multi-view captures, and smartphone-captured videos is presented, which significantly outperforms existing approaches in both monocular and multi-view settings.
- Learning Visual Generative Priors without Text8
This work develops a pure vision-based training framework, Lumos, and confirms the feasibility and the scalability of learning I2I models, and finds that, as an upstream task of T2I, the I2I model serves as a more foundational visual prior and achieves on-par or better performance than existing T2I models using only 1/10 text-image pairs for fine-tuning.
- TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification8
This work proposes an embarrassingly simple approach to better align image and text features with no need of additional data formats other than image-text pairs and indicates that attribute supervision makes vision-language models accurately localize attribute-specified objects.
- 8
- DiffDoctor: Diagnosing Image Diffusion Models Before Treating7
DiffDoctor is proposed, a twostage pipeline to assist image diffusion models in generating fewer artifacts by developing a robust artifact detector and an efficient human-in-the-loop annotation process, incorporating a carefully designed class-balance strategy.
- ScaleLSD: Scalable Deep Line Segment Detection Streamlined7
This paper studies the problem of Line Segment Detection (LSD) for the characterization of line geometry in images and proposes a domain-agnostic robust LSD model that works well for any natural images, observed to be the first deep approach that outperforms the pioneered non-deep LSD in all aspects.
- BerfScene: Bev-conditioned Equivariant Radiance Fields for Infinite 3D Scene Generation7
A practical and efficient 3D representation that incorporates an equivariant radiance field with the guidance of a bird's-eye view (BEV) map is proposed that allows for large-scale, even infinite-scale, 3D scenes via synthesizing local scenes and then stitching them with smooth consistency.
- Benchmarking and Analyzing 3D-aware Image Synthesis with a Modularized Codebase7
A well-structured codebase is built, dubbed Carver, through modularizing the generation process, which allows researchers to develop and replace each module independently, and hence offers an opportunity to fairly compare various approaches and recognize their contributions from the module perspective.
- 6
- 6
- 6
- CoDance: An Unbind-Rebind Paradigm for Robust Multi-Subject Animation5
CoDance is proposed, a novel Unbind-Rebind framework that enables the animation of arbitrary subject counts, types, and spatial configurations conditioned on a single, potentially misaligned pose sequence and achieves SOTA performance.
- 5
- ZeroDexGrasp: Zero-Shot Task-Oriented Dexterous Grasp Synthesis with Prompt-Based Multi-Stage Semantic Reasoning5
Experimental results demonstrate that ZeroDexGrasp enables high-quality zero-shot dexterous grasping on diverse unseen object categories and complex task requirements, advancing toward more generalizable and intelligent robotic grasping.
- Learning Naturally Aggregated Appearance for Efficient 3D Editing4
This work proposes to learn the color field as an explicit 2D appearance aggregation, also called canonical image, with which users can easily customize their 3D editing via 2D image processing, and complement the canonical image with a projection field that maps 3D points onto 2D pixels for texture query.
- 4
- Feature Matters: A Stage-by-Stage Approach for Knowledge Transfer.4
It is argued that feature is the most important knowledge from teacher, and sufficient for student to achieve appealing performance by just learning similar features as teacher without any processing, and presented an efficient learning strategy, which is to make student mimic features of teacher stage by stage.
- 4
- 3
- UrbanGen: Urban Generation with Compositional and Controllable Neural Fields3
The systematic study shows that UrbanGen outperforms state-of-the-art generative radiance field baselines in terms of image fidelity and geometry accuracy for urban scene generation and brings a new set of controllability features, including large camera movements, stuff editing, and city style control.
- Spatial Steerability of GANs via Self-Supervision from Discriminator3
This work designs randomly sampled Gaussian heatmaps to be encoded into the intermediate layers of generative models as spatial inductive bias, and incorporates DragGAN into the framework, which facilitates fine-grained manipulation within a reasonable time and supports a coarse-to-fine editing process.
- GenDeF: Learning Generative Deformation Field for Video Generation3
This work proposes to render a video by warping one static image with a generative deformation field (GenDeF), alleviating the difficulty in producing a video and thereby resulting in better visual quality.
- ViM: Vision Middleware for Unified Downstream Transferring3
Vision Middleware is presented, a new learning paradigm that targets unified transferring from a single foundation model to a variety of downstream tasks and can include as many midstream tasks as possible, narrowing the task gap between upstream and downstream.
- Lipschitz Singularities in Diffusion Models3
A novel approach is proposed, dubbed E-TSDM, which eliminates the Lipschitz singularity of the diffusion model near zero, and yields a substantial improvement in performance, e.g., on the high-resolution FFHQ dataset.
- GH-Feat: Learning Versatile Generative Hierarchical Features From GANs3
This work investigates that a generative feature learned from image synthesis exhibits great potentials in solving a wide range of computer vision tasks, including both generative ones and more importantly discriminative ones, and develops GH-Feat, a hierarchical visual feature with multi-level semantics.
- 3
- Efficient Object Reconstruction with Differentiable Area Light Shading2
This work proposes an inverse rendering method that uses active area lighting: as this provides a wider range of lighting angles per photo than point lighting, material reconstruction can be more accurate for the same number of photos, and improves material reconstruction, particularly for material roughness.
- Interacted Planes Reveal 3D Line Mapping1
LiP-Map pioneers the integration of planar topology into 3D line mapping, not by imposing pairwise coplanarity constraints but by explicitly constructing interactions between plane and line primitives, thus offering a principled route toward structured reconstruction in man-made environments.
- 1
- ScenDi: 3D-to-2D Scene Diffusion Cascades for Urban Generation1
ScenDi, a method for urban scene generation that integrates both 3D and 2D diffusion models that generates desired scenes based on input conditions and successfully adheres to accurate camera trajectories is proposed.
- NeuMesh++: Toward Versatile and Efficient Volumetric Editing With Disentangled Neural Mesh-Based Implicit Field1
A novel mesh-based representation by encoding the neural radiance field with disentangled geometry, texture, and semantic codes on mesh vertices is presented, which empowers a set of efficient and comprehensive editing functionalities, including mesh-guided geometry editing, designated texture editing with texture swapping, filling and painting operations, and semantic-guided editing.
- 1
- 1
- HeadArtist-VL: Vision / Language Guided 3D Head Generation with Self Score Distillation1
In the experiments, the HeadArtist-VL produces high-quality 3D head sculptures with rich geometry and photo-realistic appearance, which significantly outperforms state-of-the-art methods.
- 1
- 1
- 1
- 1
- 1
- Interpreting Class Conditional GANs with Channel Awareness1
Channel awareness is proposed, which quantitatively characterizes how a single channel contributes to the final synthesis of multiple classes in the class conditional generator, and shows promising potential in segmenting the synthesized image and evaluating the category-wise synthesis performance.
- 1
- 1
- Application of Principal Components Analysis and Cluster Analysis in Evaluation of Blending Module1
The results showed that the multivariate analysis overcome the shortcomings of traditional physical, chemical and sensory index tests which failed to comprehensively evaluate and classify between numerous tobacco samples.
- –
- Study on the evaluation method of cigarette astringency in the simulated oral environment–
The study concluded that coefficient of friction was a more reliable measure for assessing the extent of astringency in cigarettes than the total phenol content and pH of MS, offering new insights into astringency evaluation and development of high-grade cigarettes.
- –
- –
- –
- A Novel Chinese Address Segmentation Method with Self-growth Feature–
This work proposes a novel CAS method or model that starts from scratch, without relying on any pre-installed knowledge about Chinese addresses, and dynamically evolves and grows its knowledge library by leveraging contextual information and comparing addresses during the process of dividing them into address elements.
- –
- Intravoxel incoherent motion diffusion-weighted imaging for monitoring the anti-angiogenesis efficacy in a C6 glioma rat model–
Investigation of the feasibility of intravoxel incoherent motion diffusion-weighted imaging in evaluating early effects of antiangiogenic therapy in a C6 glioma rat model showed significantly decreased D* values and increased ADC and D values corresponding to histological staining in Bev treated tumors.
- –
- –
- –
- Neural Dependencies Emerging from Learning Massive Categories–
It is demonstrated that identifying neural dependencies is equivalent to solving the Covariance Lasso (CovLasso) regression problem proposed in this paper, and the potential of neural dependencies in understanding internal data correlations, generalizing models to unseen categories, and improving model robustness with a dependency-derived regularizer is shown.
- –
- –
- Research Progress of Nursing Safety Management of Pre-hospital Care–
This article reviewed the implementation of domestic hospital nursing safety and specific management in pre-hospital care in recent years, and stated current situation of nursing safety management inPre- hospital care and its practical application prospect.
- –
- –
- Statistical Technique in Quality Management of Military Products and Application of SPSS Software–
The paper introduces the simple and practical SPSS software into quality management of military products so as to simplify the difficulty of application of statistical technique.
- –
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.