Jae-Pil Heo

16

Papers

309

Total Citations

Papers (16)

Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer

VLCounter: Text-Aware Visual Representation for Zero-Shot Object Counting

Task-Disruptive Background Suppression for Few-Shot Segmentation

Fine-Tuning Visual Autogressive Models for Subject-Driven Generation

Diversity-aware Channel Pruning for StyleGAN Compression

Prediction-Feedback DETR for Temporal Action Detection

Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot Segmentation

Ambiguity-Restrained Text-Video Representation Learning for Partially Relevant Video Retrieval

Auto-Encoded Supervision for Perceptual Image Super-Resolution

Temporal Alignment-Free Video Matching for Few-shot Action Recognition

Noise-Free Optimization in Early Training Steps for Image Super-resolution

Towards Squeezing-Averse Virtual Try-On via Sequential Deformation

Selective Contrastive Learning for Weakly Supervised Affordance Grounding

Prototypes are Balanced Units for Efficient and Effective Partially Relevant Video Retrieval

Translation of Text Embedding via Delta Vector to Suppress Strongly Entangled Content in Text-to-Image Diffusion Models

Bridging the Semantic Granularity Gap Between Text and Frame Representations for Partially Relevant Video Retrieval