We tackle the common challenge of inter-concept visual confusion in compositional concept generation using text-guided diffusion models (TGDMs). It becomes even more pronounced in the generation of customized concepts, due to the scarcity of user-provided concept visual examples. By revisiting the two major stages leading to the success of TGDMs---1) contrastive image-language pre-training (CLIP) for text encoder that encodes visual semantics, and 2) training TGDM that decodes the textual embeddings into pixels---we point that existing customized generation methods only focus on fine-tuning the second stage while overlooking the first one. To this end, we propose a simple yet effective solution called CLIF: contrastive image-language fine-tuning. Specifically, given a few samples of customized concepts, we obtain non-confusing textual embeddings of a concept by fine-tuning CLIP via contrasting a concept and the over-segmented visual regions of other concepts. Experimental results demonstrate the effectiveness of CLIF in preventing the confusion of multi-customized concept generation.
Concpet Bank. We curate a dataset consisting of $18$ representative characters, including 9 real-world, 4 3D-animated, and 5 2D-animated. Each of them possesses unique visual appearances that must be preserved in the customized generation.
Pipeline of training data curation. We mix the customized concepts and common concepts at instance-level and segmentation-level, to help decoupling multi-concept token embeddings which can eliminate the confusion issues.
