SOTAVerified

Image Matting

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

Papers

Showing 110 of 225 papers

TitleStatusHype
Post-Training Quantization for Video Matting0
MP-Mat: A 3D-and-Instance-Aware Human Matting and Editing Framework with Multiplane RepresentationCode0
MaSS13K: A Matting-level Semantic Segmentation BenchmarkCode2
VRMDiff: Text-Guided Video Referring Matting Generation of DiffusionCode1
Path-Adaptive Matting for Efficient Inference Under Various Computational Cost Constraints0
Object-Aware Video Matting with Cross-Frame Guidance0
Enhancing Image Matting in Real-World Scenes with Mask-Guided Iterative Refinement0
Efficient Portrait Matte Creation With Layer Diffusion and Connectivity Priors0
MatAnyone: Stable Video Matting with Consistent Memory Propagation0
BEN: Using Confidence-Guided Matting for Dichotomous Image SegmentationCode0
Show:102550
← PrevPage 1 of 23Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1CAMMSE4.5Unverified
2BMMSE1.33Unverified
3IMMSE1.16Unverified
4Adobe LS-GANMSE0.97Unverified