Zobrazeno 1 - 10
of 211
pro vyhledávání: '"Daochang An"'
Diffusion models, widely used for image and video generation, face a significant limitation: the risk of memorizing and reproducing training data during inference, potentially generating unauthorized copyrighted content. While prior research has focu
Externí odkaz:
http://arxiv.org/abs/2410.21669
In this paper, we identify and leverage a novel `bright ending' (BE) anomaly in diffusion models prone to memorizing training images to address a new task: locating localized memorization regions within these models. BE refers to a distinct cross-att
Externí odkaz:
http://arxiv.org/abs/2410.21665
We found that enforcing guidance throughout the sampling process is often counterproductive due to the model-fitting issue, where samples are 'tuned' to match the classifier's parameters rather than generalizing the expected condition. This work iden
Externí odkaz:
http://arxiv.org/abs/2408.11194
Surgical triplet recognition is an essential building block to enable next-generation context-aware operating rooms. The goal is to identify the combinations of instruments, verbs, and targets presented in surgical video frames. In this paper, we pro
Externí odkaz:
http://arxiv.org/abs/2406.13210
Pretrained diffusion models and their outputs are widely accessible due to their exceptional capacity for synthesizing high-quality images and their open-source nature. The users, however, may face litigation risks owing to the models' tendency to me
Externí odkaz:
http://arxiv.org/abs/2404.00922
Recent advancements in generative AI have suggested that by taking visual prompt, GPT-4V can demonstrate significant proficiency in image recognition task. Despite its impressive capabilities, the financial cost associated with GPT-4V's inference pre
Externí odkaz:
http://arxiv.org/abs/2403.11468
Diffusion Probabilistic Models (DPMs) have demonstrated substantial promise in image generation tasks but heavily rely on the availability of large amounts of training data. Previous works, like GANs, have tackled the limited data problem by transfer
Externí odkaz:
http://arxiv.org/abs/2308.11948
Diffusion probabilistic models (DPMs) have been shown to generate high-quality images without the need for delicate adversarial training. However, the current sampling process in DPMs is prone to violent shaking. In this paper, we present a novel rev
Externí odkaz:
http://arxiv.org/abs/2308.11941
Publikováno v:
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 18, Pp 206-220 (2025)
Deep learning methods have achieved huge success in synthetic aperture radar (SAR) target recognition, yet insufficient data limits their performances. If the key prior knowledge that describes the target is incorporated into the network, this proble
Externí odkaz:
https://doaj.org/article/61dd58ae8a7b445fac4957816f0e8938
Temporal action segmentation is crucial for understanding long-form videos. Previous works on this task commonly adopt an iterative refinement paradigm by using multi-stage models. We propose a novel framework via denoising diffusion models, which no
Externí odkaz:
http://arxiv.org/abs/2303.17959