Publications by Rita Cucchiara

Denoising Diffusion Probabilistic Models have shown an impressive generation quality although their long sampling chain leads to high computational costs. In this paper, we observe that a long sampling chain also leads to an error accumulation phenomenon, which is similar to the exposure bias problem in autoregressive text generation. Specifically, we note that there is a discrepancy between training and testing, since the former is conditioned on the ground truth samples, while the latter is conditioned on the previously generated results. To alleviate this problem, we propose a very simple but effective training regularization, consisting in perturbing the ground truth samples to simulate the inference time prediction errors. We empirically show that, without affecting the recall and precision, the proposed input perturbation leads to a significant improvement in the sample quality while reducing both the training and the inference times. For instance, on CelebA 64×64, we achieve a new state-of-the-art FID score of 1.27, while saving 37.5% of the training time. The code is available at https://github.com/forever208/DDPM-IP.

The rapidly evolving fields of e-commerce and metaverse continue to seek innovative approaches to enhance the consumer experience. At the same time, recent advancements in the development of diffusion models have enabled generative networks to create remarkably realistic images. In this context, image-based virtual try-on, which consists in generating a novel image of a target model wearing a given in-shop garment, has yet to capitalize on the potential of these powerful generative solutions. This work introduces LaDI-VTON, the first Latent Diffusion textual Inversion-enhanced model for the Virtual Try-ON task. The proposed architecture relies on a latent diffusion model extended with a novel additional autoencoder module that exploits learnable skip connections to enhance the generation process preserving the model's characteristics. To effectively maintain the texture and details of the in-shop garment, we propose a textual inversion component that can map the visual features of the garment to the CLIP token embedding space and thus generate a set of pseudo-word token embeddings capable of conditioning the generation process. Experimental results on Dress Code and VITON-HD datasets demonstrate that our approach outperforms the competitors by a consistent margin, achieving a significant milestone for the task.

In line with current developments in indirect intergroup contact literature, we conducted a field study using the imagined contact paradigm among high-status (Italian children) and low-status (children with foreign origins) group members (N = 122; 53 females, mean age = 7.52 years). The experiment aimed to improve attitudes and behavior toward a different low-status group, children with disability. To assess behavior, we focused on an objective measure that captures the physical distance between participants and a child with disability over the course of a five-minute interaction (i.e., while playing together). Results from a 3-week intervention revealed that in the case of high-status children imagined contact, relative to a no-intervention control condition, improved outgroup attitudes and behavior, and strengthened helping and contact intentions. These effects however did not emerge among low-status children. The results are discussed in the context of intergroup contact literature, with emphasis on the implications of imagined contact for educational settings.

Fashion illustration is used by designers to communicate their vision and to bring the design idea from conceptualization to realization, showing how clothes interact with the human body. In this context, computer vision can thus be used to improve the fashion design process. Differently from previous works that mainly focused on the virtual try-on of garments, we propose the task of multimodal-conditioned fashion image editing, guiding the generation of human-centric fashion images by following multimodal prompts, such as text, human body poses, and garment sketches. We tackle this problem by proposing a new architecture based on latent diffusion models, an approach that has not been used before in the fashion domain. Given the lack of existing datasets suitable for the task, we also extend two existing fashion datasets, namely Dress Code and VITON-HD, with multimodal annotations collected in a semi-automatic manner. Experimental results on these new datasets demonstrate the effectiveness of our proposal, both in terms of realism and coherence with the given multimodal inputs. Source code and collected multimodal annotations are publicly available at: https://github.com/aimagelab/multimodal-garment-designer.

The inexorable growth of online shopping and e-commerce demands scalable and robust machine learning-based solutions to accommodate customer requirements. In the context of automatic tagging classification and multimodal retrieval, prior works either defined a low generalizable supervised learning approach or more reusable CLIP-based techniques while, however, training on closed source data. In this work, we propose OpenFashionCLIP, a vision-and-language contrastive learning method that only adopts open-source fashion data stemming from diverse domains, and characterized by varying degrees of specificity. Our approach is extensively validated across several tasks and benchmarks, and experimental results highlight a significant out-of-domain generalization capability and consistent improvements over state-of-the-art methods both in terms of accuracy and recall. Source code and trained models are publicly available at: https://github.com/aimagelab/open-fashion-clip.

The CLIP model has been recently proven to be very effective for a variety of cross-modal tasks, including the evaluation of captions generated from vision-and-language models. In this paper, we propose a new recipe for a contrastive-based evaluation metric for image captioning, namely Positive-Augmented Contrastive learning Score (PAC-S), that in a novel way unifies the learning of a contrastive visual-semantic space with the addition of generated images and text on curated data. Experiments spanning several datasets demonstrate that our new metric achieves the highest correlation with human judgments on both images and videos, outperforming existing reference-based metrics like CIDEr and SPICE and reference-free metrics like CLIP-Score. Finally, we test the system-level correlation of the proposed metric when considering popular image captioning approaches, and assess the impact of employing different cross-modal features. We publicly release our source code and trained models.

Background and objective: The functions of an organism and its biological processes result from the expression of genes and proteins. Therefore quantifying and predicting mRNA and protein levels is a crucial aspect of scientific research. Concerning the prediction of mRNA levels, the available approaches use the sequence upstream and downstream of the Transcription Start Site (TSS) as input to neural networks. The State-of-the-art models (e.g., Xpresso and Basenjii) predict mRNA levels exploiting Convolutional (CNN) or Long Short Term Memory (LSTM) Networks. However, CNN prediction depends on convolutional kernel size, and LSTM suffers from capturing long-range dependencies in the sequence. Concerning the prediction of protein levels, as far as we know, there is no model for predicting protein levels by exploiting the gene or protein sequences. Methods: Here, we exploit a new model type (called Perceiver) for mRNA and protein level prediction, exploiting a Transformer-based architecture with an attention module to attend to long-range interactions in the sequences. In addition, the Perceiver model overcomes the quadratic complexity of the standard Transformer architectures. This work's contributions are 1. DNAPerceiver model to predict mRNA levels from the sequence upstream and downstream of the TSS; 2. ProteinPerceiver model to predict protein levels from the protein sequence; 3. Protein&DNAPerceiver model to predict protein levels from TSS and protein sequences. Results: The models are evaluated on cell lines, mice, glioblastoma, and lung cancer tissues. The results show the effectiveness of the Perceiver-type models in predicting mRNA and protein levels. Conclusions: This paper presents a Perceiver architecture for mRNA and protein level prediction. In the future, inserting regulatory and epigenetic information into the model could improve mRNA and protein level predictions. The source code is freely available at https://github.com/MatteoStefanini/DNAPerceiver.

Image captioning is a challenging task that combines Computer Vision and Natural Language Processing to generate descriptive and accurate textual descriptions for input images. Research efforts in this field mainly focus on developing novel architectural components to extend image captioning models and using large-scale image-text datasets crawled from the web to boost final performance. In this work, we explore an alternative to web-crawled data and augment the training dataset with synthetic images generated by a latent diffusion model. In particular, we propose a simple yet effective synthetic data augmentation framework that is capable of significantly improving the quality of captions generated by a standard Transformer-based model, leading to competitive results on the COCO dataset.