HairCLIP: Design Your Hair by Text and Reference Image

ITstyler: Image-optimized Text-based Style Transfer

“…Nevertheless, it still heavily relies on pre-trained generative models. HairCLIP [23] achieves disentangled hair editing by feeding separate hairstyle and hair color information into different sub hair mappers to map the input conditions into corresponding latent code changes.…”

Clip-Rr: Improved Clip Network for Relation-Focused Cross-Modal Information Retrieval

“…However, due to the advantage of ViT on pre-training, the ViT encoder outperformed the CNN encoder, and it is most commonly applied in other works [9]. CLIP has recently been used in various tasks, such as e-commerce image retrieval [9], text-image generation [10], and image segmentation [18]. However, CLIP still lacks the ability to effectively match local information in images to their descriptions in cross-modal information retrieval tasks [12].…”

“…Recently, Radford et al [8] proposed a Contrastive Language-Image Pre-training network (CLIP) which achieved state-of-the-art performance in cross-modal tasks. CLIP employs the pre-trained GPT-2 and Visual Transformer (ViT) to encode descriptions and images into a shared embedding space respectively [8], and has been widely applied to various information retrieval tasks such as e-commerce image retrieval [9], and text-image generation [10]. One of CLIP's limitations is that it cannot identify relations between objects in images.…”

TextIR: A Simple Framework for Text-based Editable Image Restoration

“…With the successful development of cross-modal visual and linguistic representations [30,42,43,54], especially the omnipotent CLIP [35], many efforts [7, 18,23,34,46,49,51] have recently started investigating text-driven image manipulation. However, there are no existing methods specifically for image restoration.…”

“…However, there are no existing methods specifically for image restoration. Among these works, the most relevant ones are StyleCLIP [34], HairCLIP [49], and CLIP-Styler [23]. StyleCLIP performs attribute manipulation with exploring learned latent space of StyleGANv2 [21].…”