Peng-Tao Jiang

Researcher@vivo Mobile Communication Co., Ltd·中国·OpenReview

研究方向

Image/Video editing · Diffusion Application · Multi-task learning · Image restoration/enhancement · Scene parsing · Visual attention · Weakly supervised semantic segmentation

Peng-Tao Jiang

Any-to-Bokeh: Arbitrary-Subject Video Refocusing with Video Diffusion Model

TS-Attn: Temporal-wise Separable Attention for Multi-Event Video Generation

Trust but Verify: Adaptive Conditioning for Reference-Based Diffusion Super-Resolution via Implicit Reference Correlation Modeling

I-DRUID: Layout to image generation via instance-disentangled representation and unpaired data

RED: Robust Event-Guided Motion Deblurring with Modality-Specific Disentangled Representation

MagicTryOn: Harnessing Diffusion Transformer for Garment-Preserving Video Virtual Try-on

HyperMotion: DiT-Based Pose-Guided Human Image Animation of Complex Motions

Diffusion-Based Photorealistic Bokeh Rendering for Mobile Devices

Beyond Fixed: Aligning Guidance with Diffusion Dynamics via Exponential Scaling

Learning Differential Pyramid Representation for Tone Mapping

Photography Perspective Composition: Towards Aesthetic Perspective Recommendation

SE-GUI: Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning

Learning Adaptive Lighting via Channel-Aware Guidance

High-Precision Dichotomous Image Segmentation via Probing Diffusion Capacity

Multi-Task Dense Predictions via Unleashing the Power of Diffusion

ClearSR: Latent Low-Resolution Image Embeddings Help Diffusion-Based Real-World Super Resolution Models See Clearer

Learning Differential Pyramid Representation for Tone Mapping

Learning Adaptive Lighting via Channel-Aware Guidance

ConsisSR: Delving Deep into Consistency in Diffusion-based Image Super-Resolution

Towards Natural Image Matting in the Wild via Real-Scenario Prior