Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Antitza DantchevaSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works68 from public data
- VBench: Comprehensive Benchmark Suite for Video Generative Models2,062
VBench is a comprehensive benchmark suite that dissects “video generation quality” into specific, hierarchical, and disentangled dimensions, each with tailored prompts and evaluation methods, and investi-gate the gaps between video and image generation models.
- AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning1,729
The proposed AnimateDiff is a practical framework for animating personalized T2I models without requiring model-specific tuning and MotionLoRA is a lightweight fine-tuning technique for AnimateDiff that enables a pre-trained motion module to adapt to new motion patterns, such as different shot types, at a low training and data collection cost.
- InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation557
A scalable approach to autonomously build a high-quality video-text dataset with large language models (LLM), thereby showcasing its efficacy in learning video-language representation at scale and introducing ViCLIP, a video- Text representation learning model based on ViT-L.
- LaVie: High-Quality Video Generation with Cascaded Latent Diffusion Models426
The incorporation of simple temporal self-attentions, coupled with rotary positional encoding, adequately captures the temporal correlations inherent in video data and proves that the process of joint image-video fine-tuning plays a pivotal role in producing high-quality and creative outcomes.
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step378
This work first derive a deterministic sampling process from the most recent state-of-the-art (SOTA) method for accelerating diffusion-based SR, which can be distilled into a student model that performs SR within only one inference step, resulting in a remarkable up to × 10 speedup for inference.
- 278
- SEINE: Short-to-Long Video Diffusion Model for Generative Transition and Prediction267
This paper proposes a random-mask video diffusion model to automatically generate transitions based on textual descriptions that generates transition videos that ensure coherence and visual quality and validate the effectiveness of the approach over existing methods for generative transition and prediction.
- Latent Image Animator: Learning to Animate Images via Latent Space Navigation234
The Latent Image Animator is introduced, a self-supervised autoencoder that evades need for structure representation and significantly outperforms state-of-art methods on VoxCeleb, Taichi and TED-talk datasets w.r.t. generated quality.
- Joint Generative and Contrastive Learning for Unsupervised Person Re-identification196
This paper incorporates a Generative Adversarial Network and a contrastive learning module into one joint training framework, and proposes a mesh-based view generator that significantly outperforms state-of-the-art methods under both, fully unsupervised and unsuper supervised domain adaptive settings on several large scale ReID dat-sets.
- ImaGINator: Conditional Spatio-Temporal GAN for Video Generation157
A novel conditional GAN architecture, namely ImaGINator, which given a single image, a condition (label of a facial expression or action) and noise, decomposes appearance and motion in both latent and high level feature spaces, generating realistic videos.
- 111
- EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion90
The proposed EpiDiff is a localized interactive multiview diffusion model to insert a lightweight epipolar attention block into the frozen diffusion model, leveraging epipolar constraints to enable cross-view interaction among feature maps of neighboring views.
- Vlogger: Make Your Dream A Vlog85
Vlogger, a generic AI system for generating a minute-level video blog of user de-scriptions, smartly leverages Large Language Model (LLM) as Director and decomposes a long video generation task of vlog into four key stages, where various foundation models play the critical roles of vlog profession-als.
- UNIK: A Unified Framework for Real-world Skeleton-based Action Recognition66
Experimental results show that the proposed UNIK, with pre-training on Posetics, generalizes well and outperforms state-of-the-art when transferred onto four target action classification datasets: Toyota Smarthome, Penn Action, NTU-RGB+D 60 andNTU- RGB+D 120.
- Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models61
This work presents Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation, and proposes a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems.
- 59
- LEO: Generative Latent Image Animator for Human Video Synthesis57
Extensive quantitative and qualitative analysis suggests that LEO significantly improves coherent synthesis of human videos over previous methods on the datasets TaichiHD, FaceForensics and CelebV-HQ, as well as content-preserving video editing.
- View-Invariant Skeleton Action Representation Learning via Motion Retargeting42
This work introduces ViA, a novel View-Invariant Autoencoder for self-supervised skeleton action representation learning that leverages motion retargeting between different human performers as a pretext task, in order to disentangle the latent action-specific ‘Motion’ features on top of the visual representation of a 2D or 3D skeleton sequence.
- LIA: Latent Image Animator41
The Latent Image Animator (LIA), streamlined to animate high-resolution images, outperforms state-of-the-art on VoxCeleb, TaichiHD, and TED-talk datasets with respect to video quality and spatio-temporal consistency.
- Learning Invariance From Generated Variance for Unsupervised Person Re-Identification41
This article proposes to replace traditional data augmentation with a generative adversarial network (GAN) that is targeted to generate augmented views for contrastive learning and achieves new state-of-the-art unsupervised person ReID performance on mainstream large-scale benchmarks.
- Ouroboros3D: Image-to-3D Generation via 3D-aware Recursive Diffusion36
Ouroboros3D, an end-to-end trainable framework that integrates multi-view generation and 3D reconstruction into a recursive diffusion process through feedback mechanism, achieves superior multi-view consistency and producing 3D models with higher geometric realism.
- Long-Term Rhythmic Video Soundtracker31
Long-Term Rhythmic Video Soundtracker (LORIS), a novel framework to synthesize long-term conditional waveforms consisting of a latent conditional diffusion probabilistic model to perform waveform synthesis, is presented.
- CineTrans: Learning to Generate Videos with Cinematic Transitions via Masked Diffusion Models29
This work introduces CineTrans, a novel framework for generating coherent multi-shot videos with cinematic, film-style transitions, and proposes specialized evaluation metrics for transition control, temporal consistency and overall quality.
- 26
- Hierarchical Diffusion Autoencoders and Disentangled Image Manipulation26
The proposed Hierarchical Diffusion Autoencoders (HDAE) that exploits the coarse-to-fine feature hierarchy for the latent space of diffusion models and proposes a truncated feature based approach for disentangled image manipulation.
- LAC - Latent Action Composition for Skeleton-based Action Segmentation24
Latent Action Composition is proposed, a novel self-supervised framework aiming at learning from synthesized composable motions for skeleton-based action segmentation and representations learned from pre-trained LAC outperform the state-of-the-art by a large margin on TSU, Charades, PKU-MMD datasets.
- The Devil is in the Prompts: Retrieval-Augmented Prompt Optimization for Text-To-Video Generation23
RAPO refines the naive prompts through dual optimization branches, selecting the superior prompt for T2V generation, demonstrating the significance of prompt optimization for user-provided prompts.
- Vinci: A Real-time Smart Assistant Based on Egocentric Vision-language Model for Portable Devices22
Vinci is a vision-language system designed to provide real-time, comprehensive AI assistance on portable devices that integrates an egocentric vision foundation model with a large language model (LLM), enabling advanced functionalities such as scene understanding, temporal grounding, video summarization, and future planning.
- From Attribute-Labels to Faces: Face Generation Using a Conditional Generative Adversarial Network22
This work proposes two models for attribute-label based facial image and video generation incorporating 2D and 3D deep conditional generative adversarial networks (DCGAN) and indicates the methods' ability to generate realistic faces from attribute labels.
- Vinci: A Real-time Embodied Smart Assistant based on Egocentric Vision-Language Model21
Vinci, a real-time embodied smart assistant built upon an egocentric vision-language model, is introduced, a real-time embodied smart assistant built upon an egocentric vision-language model designed for deployment on portable devices such as smartphones and wearable cameras.
- 20
- Selective Spatio-Temporal Aggregation Based Pose Refinement System: Towards Understanding Human Activities in Real-World Videos18
A Selective Spatio-Temporal Aggregation mechanism that refines and smooths the keypoint locations extracted by multiple expert pose estimators, and an effective weakly-supervised self-training framework which leverages the aggregated poses as pseudo ground-truth in-stead of handcrafted annotations for real-world pose estimation are proposed.
- 18
- Cinemo: Consistent and Controllable Image Animation with Motion Diffusion Models16
This paper introduces Cinemo, a novel image animation approach towards achieving better motion controllability, as well as stronger temporal consistency and smoothness, and proposes three effective strategies at the training and inference stages of Cinemo to accomplish this goal.
- 16
- 13
- 13
- 13
- 12
- RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling11
A cross-stage prompt optimization framework that unifies training-data--aligned refinement, test-time iterative scaling, and large language model (LLM) fine-tuning to substantially improve T2V generation without modifying the underlying generative backbone is presented.
- 11
- An Egocentric Vision-Language Model based Portable Real-time Smart Assistant9
Vinci is a vision-language system designed to provide real-time, comprehensive AI assistance on portable devices that integrates an egocentric vision foundation model with a large language model (LLM), enabling advanced functionalities such as scene understanding, temporal grounding, video summarization, and future planning.
- Comparing Methods for Assessment of Facial Dynamics in Patients with Major Neurocognitive Disorders9
This work compares methods for assessing facial dynamics such as talking, singing, neutral and smiling in AD-patients, captured during music mnemotherapy sessions, and suggests that the methods generalize well to the context of facial dynamics.
- GenHOI: Generalizing Text-driven 4D Human-Object Interaction Synthesis for Unseen Objects8
GenHOI is introduced, a novel two-stage framework aimed at achieving two key objectives: 1) generalization to unseen objects and 2) the synthesis of high-fidelity 4D HOI sequences.
- MuLan: Adapting Multilingual Diffusion Models for Hundreds of Languages with Negligible Cost8
This work introduces MuLan, Multi-Language adapter, a lightweight language adapter with fewer than 20M parameters, trained alongside a frozen text encoder and image diffusion model, to explore a cost-effective framework for multilingual image generation.
- 7
- ConditionVideo: Training-Free Condition-Guided Text-to-Video Generation7
This work introduces ConditionVideo, a training-free approach to text-to-video generation based on the provided condition, video, and input text, by leveraging the power of off-the-shelf text- to-image generation methods (e.g., Stable Diffusion).
- Dimitra: Audio-driven Diffusion model for Expressive Talking Head Generation6
Quantitative and qualitative experiments showcase that Dimitra is able to outperform existing approaches for generating realistic talking heads imparting lip motion, facial expression, and head pose, and head pose realism.
- 6
- CausalMotion: Structured Physical Reasoning as Keyframe and Trajectory Guidance for Training-Free Video Generation5
The key idea is to decouple reasoning from generation by leveraging a vision-language model to decompose a text prompt into a sequence of causally consistent keyframes and object-centric motion trajectories that guide a pretrained video diffusion model during inference.
- ShotDirector: Directorially Controllable Multi-Shot Video Generation with Cinematographic Transitions5
This work proposes ShotDirector, an efficient framework that integrates parameter-level camera control and hierarchical editing-pattern-aware prompting that effectively combines parameter-level conditions with high-level semantic guidance, achieving film-like controllable shot transitions.
- VDOT: Efficient Unified Video Creation via Optimal Transport Distillation5
This work proposes an efficient unified video creation model, named VDOT, that integrates a discriminator to enable the model to perceive real video data, thereby enhancing the quality of generated videos and curating a unified testing benchmark, UVCBench, to standardize evaluation.
- G 3 AN: This video does not exist. Disentangling motion and appearance for video generation.5
The novel spatio-temporal GAN-architecture $G^3AN$, which seeks to capture the distribution of high dimensional video data and to model appearance and motion in disentangled manner, outperforms state-of-the-art methods on the facial expression datasets MUG and UvA-NEMO, as well as the Weizmann and UCF101 datasets on human action.
- 4
- 4
- Consistent and Controllable Image Animation with Motion Linear Diffusion Transformers3
MiraMo is a framework designed to enhance efficiency, appearance consistency, and motion smoothness in image animation and validate the superiority of MiraMo in generating consistent, smooth, and controllable animations with accelerated inference speed.
- Training-free Stylized Text-to-Image Generation with Fast Inference3
A novel stylized image generation method leveraging a pre-trained large-scale diffusion model without requiring fine-tuning or any additional optimization, termed as OmniPainter, which exploits the self-consistency property of latent consistency models to extract the representative style statistics from reference style images to guide the stylization process.
- AI killed the Video Star. Audio-Driven Diffusion Model for Expressive Talking Head Generation2
Quantitative and qualitative experiments, as well as a user study on two widely employed datasets, suggest that Dimitra++ is able to outperform existing approaches in generating realistic talking heads imparting lip motion, facial expression, and head pose.
- Diff-VF: Training-free High-quality Long Video Generation via Diffusion Model2
Diff-VF, a training-free, plug-and-play and model-agnostic framework that converts existing short-video diffusion backbones into long-video generators without modifying or fine-tuning the base model, is proposed.
- PARE: Pruning and Adaptive Routing for Efficient Video Generation1
Pruning and Adaptive Routing for Efficient video generation is proposed, which jointly compresses width and depth with structure-aware pruning and input-adaptive routing and composes with step distillation for further acceleration.
- 1
- 1
- Policy-as-Data: Learning Generalizable HOI Diffusion Models from Simulated Physics–
A novel framework that leverages a physics simulator to overcome the data-scarcity bottleneck in HOI generation and introduces a coarse-to-fine retargeting process that bridges the representation gap between the simplified model used in physics simulator and the standard parametric body models required for generative training.
- –
- –
- –
- –
- –
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.