Lightweight Portrait Matting via Regional Attention and Refinement
원문 초록
We present a lightweight model for high resolution portrait matting. The model does not use any auxiliary inputs such as trimaps or background captures and achieves real time performance for HD videos and near real time for 4K. Our model is built upon a two-stage framework with a low resolution network for coarse alpha estimation followed by a refinement network for local region improvement. However, a naive implementation of the two-stage model suffers from poor matting quality if not utilizing any auxiliary inputs. We address the performance gap by leveraging the vision transformer (ViT) as the backbone of the low resolution network, motivated by the observation that the tokenization step of ViT can reduce spatial resolution while retain as much pixel information as possible. To inform local regions of the context, we propose a novel cross region attention (CRA) module in the refinement network to propagate the contextual information across the neighboring regions. We demonstrate that our method achieves superior results and outperforms other baselines on three benchmark datasets while only uses $1/20$ of the FLOPS compared to the existing state-of-the-art model.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
MaSS13K: A Matting-level Semantic Segmentation Benchmark
Dynamic Parameter Memory: Temporary LoRA-Enhanced LLM for Long-Sequence Emotion Recognition in Conversation
ComfyUI-R1: Exploring Reasoning Models for Workflow Generation
SeerAttention-R: Sparse Attention Adaptation for Long Reasoning
Unfolding Spatial Cognition: Evaluating Multimodal Models on Visual Simulations
Ultra-High-Resolution Image Synthesis: Data, Method and Evaluation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.