TP2O: Creative Text Pair-to-Object Generation Using Balance Swap-Sampling

J Li, Z Zhang, J Yang - European Conference on Computer Vision, 2025 - Springer
Generating creative combinatorial objects from two seemingly unrelated object texts is a
challenging task in text-to-image synthesis, often hindered by a focus on emulating existing …

Diffusion models as artists: are we closing the gap between humans and machines?

V Boutin, T Fel, L Singhal, R Mukherji… - arXiv preprint arXiv …, 2023 - arxiv.org
An important milestone for AI is the development of algorithms that can produce drawings
that are indistinguishable from those of humans. Here, we adapt the'diversity vs …

[HTML][HTML] Coarse-to-fine structure and semantic learning for single-sample SAR image generation

X Wang, B Hui, P Guo, R Jin, L Ding - Remote Sensing, 2024 - mdpi.com
Synthetic Aperture Radar (SAR) enables the acquisition of high-resolution imagery even
under severe meteorological and illumination conditions. Its utility is evident across a …

Just say the name: Online continual learning with category names only via data generation

M Seo, S Cho, M Lee, D Misra, H Choi, SJ Kim… - arXiv preprint arXiv …, 2024 - arxiv.org
Requiring extensive human supervision is often impractical for continual learning due to its
cost, leading to the emergence of'name-only continual learning'that only provides the name …

Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks

V Boutin, R Mukherji, A Agrawal, S Muzellec… - arXiv preprint arXiv …, 2024 - arxiv.org
Humans can effortlessly draw new categories from a single exemplar, a feat that has long
posed a challenge for generative models. However, this gap has started to close with recent …

Dual Thinking and Perceptual Analysis of Deep Learning Models using Human Adversarial Examples

K Dayanandan, A Sinha, B Lall - arXiv preprint arXiv:2406.06967, 2024 - arxiv.org
The dual thinking framework considers fast, intuitive processing and slower, logical
processing. The perception of dual thinking in vision requires images where inferences from …

Abstracted Gaussian Prototypes for One-Shot Concept Learning

C Zou, KJ Kurtz - arXiv preprint arXiv:2408.17251, 2024 - arxiv.org
We introduce a cluster-based generative image segmentation framework to encode higher-
level representations of visual concepts based on one-shot learning inspired by the …

Top-down generation of low-resolution representations improves visual perception and imagination

Z Bi, H Li, L Tian - Neural Networks, 2024 - Elsevier
Perception or imagination requires top-down signals from high-level cortex to primary visual
cortex (V1) to reconstruct or simulate the representations bottom-up stimulated by the seen …