https://github.com/jacksyu/hoi-matting
공식 코드Virtual Multi-Modality Self-Supervised Foreground Matting for Human-Object Interaction
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Image Matting is the process of accurately estimating the foreground object in images and videos. It is a very important technique in image and video editing applications, particularly in film production for creating visual effects. In case of image segmentation, we segment the image into foreground and background by labeling the pixels. Image segmentation generates a binary image, in which a pixel either belongs to foreground or background. However, Image Matting is different from the image segmentation, wherein some pixels may belong to foreground as well as background, such pixels are called partial or mixed pixels. In order to fully separate the foreground from the background in an image, accurate estimation of the alpha values for partial or mixed pixels is necessary.
Source: Automatic Trimap Generation for Image Matting
Image Source: Real-Time High-Resolution Background Matting
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Virtual Multi-Modality Self-Supervised Foreground Matting for Human-Object Interaction
DFIMat: Decoupled Flexible Interactive Matting in Multi-Person Scenarios
Castle in the Sky: Dynamic Sky Replacement and Harmonization in Videos
JIT-Masker: Efficient Online Distillation for Background Matting
Improving Deep Image Matting via Local Smoothness Assumption
GIMP-ML: Python Plugins for using Computer Vision Models in GIMP
CAP-VSTNet: Content Affinity Preserved Versatile Style Transfer
Deep Video Matting via Spatio-Temporal Alignment and Aggregation
Human Instance Matting via Mutual Guidance and Multi-Instance Refinement
FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures
SAPA: Similarity-Aware Point Affiliation for Feature Upsampling
RenderMe-360: A Large Digital Asset Library and Benchmarks Towards High-fidelity Head Avatars
Towards Ghost-free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GAN
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.