Transformer-based visual segmentation: A survey
Visual segmentation seeks to partition images, video frames, or point clouds into multiple
segments or groups. This technique has numerous real-world applications, such as …
segments or groups. This technique has numerous real-world applications, such as …
Deep image matting: A comprehensive survey
Image matting refers to extracting precise alpha matte from natural images, and it plays a
critical role in various downstream applications, such as image editing. Despite being an ill …
critical role in various downstream applications, such as image editing. Despite being an ill …
Vitmatte: Boosting image matting with pre-trained plain vision transformers
Image matting is an inverse fusion process that separates the foreground and background
information by predicting alpha matte for each pixel. Recently, plain vision Transformers …
information by predicting alpha matte for each pixel. Recently, plain vision Transformers …
Treating pseudo-labels generation as image matting for weakly supervised semantic segmentation
Generating accurate pseudo-labels under the supervision of image categories is a crucial
step in Weakly Supervised Semantic Segmentation (WSSS). In this work, we propose a Mat …
step in Weakly Supervised Semantic Segmentation (WSSS). In this work, we propose a Mat …
Matte anything: Interactive natural image matting with segment anything model
Natural image matting algorithms aim to predict the transparency map (alpha-matte) with the
trimap guidance. However, the production of trimap often requires significant labor, which …
trimap guidance. However, the production of trimap often requires significant labor, which …
X-paste: Revisiting scalable copy-paste for instance segmentation using clip and stablediffusion
Copy-Paste is a simple and effective data augmentation strategy for instance segmentation.
By randomly pasting object instances onto new background images, it creates new training …
By randomly pasting object instances onto new background images, it creates new training …
Cle diffusion: Controllable light enhancement diffusion model
Low light enhancement has gained increasing importance with the rapid development of
visual creation and editing. However, most existing enhancement algorithms are designed …
visual creation and editing. However, most existing enhancement algorithms are designed …
Diffusion for natural image matting
Existing natural image matting algorithms inevitably have flaws in their predictions on
difficult cases, and their one-step prediction manner cannot further correct these errors. In …
difficult cases, and their one-step prediction manner cannot further correct these errors. In …