LIVIVO - The Search Portal for Life Sciences

zur deutschen Oberfläche wechseln
Advanced search

Search results

Result 1 - 6 of total 6

Search options

  1. Book ; Online: Video ControlNet

    Chu, Ernie / Lin, Shuo-Yen / Chen, Jun-Cheng

    Towards Temporally Consistent Synthetic-to-Real Video Translation Using Conditional Image Diffusion Models

    2023  

    Abstract: In this study, we present an efficient and effective approach for achieving temporally consistent synthetic-to-real video translation in videos of varying lengths. Our method leverages off-the-shelf conditional image diffusion models, allowing us to ... ...

    Abstract In this study, we present an efficient and effective approach for achieving temporally consistent synthetic-to-real video translation in videos of varying lengths. Our method leverages off-the-shelf conditional image diffusion models, allowing us to perform multiple synthetic-to-real image generations in parallel. By utilizing the available optical flow information from the synthetic videos, our approach seamlessly enforces temporal consistency among corresponding pixels across frames. This is achieved through joint noise optimization, effectively minimizing spatial and temporal discrepancies. To the best of our knowledge, our proposed method is the first to accomplish diverse and temporally consistent synthetic-to-real video translation using conditional image diffusion models. Furthermore, our approach does not require any training or fine-tuning of the diffusion models. Extensive experiments conducted on various benchmarks for synthetic-to-real video translation demonstrate the effectiveness of our approach, both quantitatively and qualitatively. Finally, we show that our method outperforms other baseline methods in terms of both temporal consistency and visual quality.
    Keywords Computer Science - Computer Vision and Pattern Recognition
    Subject code 006
    Publishing date 2023-05-30
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  2. Book ; Online: MeDM

    Chu, Ernie / Huang, Tzuhsuan / Lin, Shuo-Yen / Chen, Jun-Cheng

    Mediating Image Diffusion Models for Video-to-Video Translation with Temporal Correspondence Guidance

    2023  

    Abstract: This study introduces an efficient and effective method, MeDM, that utilizes pre-trained image Diffusion Models for video-to-video translation with consistent temporal flow. The proposed framework can render videos from scene position information, such ... ...

    Abstract This study introduces an efficient and effective method, MeDM, that utilizes pre-trained image Diffusion Models for video-to-video translation with consistent temporal flow. The proposed framework can render videos from scene position information, such as a normal G-buffer, or perform text-guided editing on videos captured in real-world scenarios. We employ explicit optical flows to construct a practical coding that enforces physical constraints on generated frames and mediates independent frame-wise scores. By leveraging this coding, maintaining temporal consistency in the generated videos can be framed as an optimization problem with a closed-form solution. To ensure compatibility with Stable Diffusion, we also suggest a workaround for modifying observation-space scores in latent Diffusion Models. Notably, MeDM does not require fine-tuning or test-time optimization of the Diffusion Models. Through extensive qualitative, quantitative, and subjective experiments on various benchmarks, the study demonstrates the effectiveness and superiority of the proposed approach. Our project page can be found at https://medm2023.github.io

    Comment: Accepted as a conference paper in AAAI 2024. Project page: https://medm2023.github.io
    Keywords Computer Science - Computer Vision and Pattern Recognition
    Subject code 004
    Publishing date 2023-08-19
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  3. Book ; Online: Diffusion to Confusion

    Lin, Shuo-Yen / Chu, Ernie / Lin, Che-Hsien / Chen, Jun-Cheng / Wang, Jia-Ching

    Naturalistic Adversarial Patch Generation Based on Diffusion Model for Object Detector

    2023  

    Abstract: Many physical adversarial patch generation methods are widely proposed to protect personal privacy from malicious monitoring using object detectors. However, they usually fail to generate satisfactory patch images in terms of both stealthiness and attack ...

    Abstract Many physical adversarial patch generation methods are widely proposed to protect personal privacy from malicious monitoring using object detectors. However, they usually fail to generate satisfactory patch images in terms of both stealthiness and attack performance without making huge efforts on careful hyperparameter tuning. To address this issue, we propose a novel naturalistic adversarial patch generation method based on the diffusion models (DM). Through sampling the optimal image from the DM model pretrained upon natural images, it allows us to stably craft high-quality and naturalistic physical adversarial patches to humans without suffering from serious mode collapse problems as other deep generative models. To the best of our knowledge, we are the first to propose DM-based naturalistic adversarial patch generation for object detectors. With extensive quantitative, qualitative, and subjective experiments, the results demonstrate the effectiveness of the proposed approach to generate better-quality and more naturalistic adversarial patches while achieving acceptable attack performance than other state-of-the-art patch generation methods. We also show various generation trade-offs under different conditions.
    Keywords Computer Science - Computer Vision and Pattern Recognition
    Subject code 004
    Publishing date 2023-07-16
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  4. Article: Evidence that dehydroepiandrosterone, DHEA, directly inhibits GnRH gene expression in GT1-7 hypothalamic neurons.

    Cui, Hong / Lin, Shuo-Yen J / Belsham, Denise D

    Molecular and cellular endocrinology

    2002  Volume 203, Issue 1-2, Page(s) 13–23

    Abstract: Dehydroepiandrosterone (DHEA) has been reported to have diverse effects on overall physiology, although its mechanism of action and specific receptor are not yet known. We have used the immortalized, clonal GT1-7 hypothalamic neurons to study DHEA ... ...

    Abstract Dehydroepiandrosterone (DHEA) has been reported to have diverse effects on overall physiology, although its mechanism of action and specific receptor are not yet known. We have used the immortalized, clonal GT1-7 hypothalamic neurons to study DHEA effects on gonadotropin-releasing hormone (GnRH) gene expression. DHEA (10(-4) M) downregulates GnRH transcription by 39, 70 and 83% at 24, 36, and 48 h, respectively, while DHEA-sulphate had no effect. Hydroxyflutamide a specific androgen receptor (AR) antagonist, and cyproterone acetate or trilostane, both inhibitors of 3 beta-hydroxysteroid dehydrogenase/delta 4,5 isomerase, the rate-limiting enzyme for the conversion of DHEA to sex steroids, did not affect the ability of DHEA to downregulate GnRH gene expression. We found that GT1-7 cells did not express aromatase, thereby precluding conversion to estrogen. Analysis of [(14)C] DHEA metabolism by thin layer chromatography indicates that the main metabolites produced are 7 alpha- and 7 beta-hydroxy DHEA, and 7-oxo DHEA, although these steroids were not able to repress GnRH gene expression alone. Cell viability studies indicated that the transcriptional repression observed is not due to GT1-7 cell death. Interestingly, SV40 T-antigen mRNA levels, under the control of 2.3 kb of the rat GnRH gene 5' regulatory region, are also repressed by DHEA. Our studies indicate that DHEA has direct effects on GnRH transcription that appear to be unique from those observed after conversion to other steroidogenic compounds.
    MeSH term(s) 5' Untranslated Regions/genetics ; Animals ; Cell Line ; Cell Survival/drug effects ; Dehydroepiandrosterone/metabolism ; Dehydroepiandrosterone/pharmacology ; Dehydroepiandrosterone/physiology ; Down-Regulation ; Estrogens/physiology ; Gonadotropin-Releasing Hormone/antagonists & inhibitors ; Gonadotropin-Releasing Hormone/genetics ; Hypothalamus/cytology ; Kinetics ; Mice ; Neurons/metabolism ; RNA, Messenger/analysis ; RNA, Messenger/drug effects ; Transcription, Genetic/drug effects
    Chemical Substances 5' Untranslated Regions ; Estrogens ; RNA, Messenger ; Gonadotropin-Releasing Hormone (33515-09-2) ; Dehydroepiandrosterone (459AG36T1B)
    Language English
    Publishing date 2002-09-04
    Publishing country Ireland
    Document type Journal Article ; Research Support, Non-U.S. Gov't
    ZDB-ID 187438-x
    ISSN 1872-8057 ; 0303-7207
    ISSN (online) 1872-8057
    ISSN 0303-7207
    DOI 10.1016/s0303-7207(03)00121-7
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  5. Article: IGF-I signaling prevents dehydroepiandrosterone (DHEA)-induced apoptosis in hypothalamic neurons.

    Lin, Shuo-Yen J / Cui, Hong / Yusta, Bernardo / Belsham, Denise D

    Molecular and cellular endocrinology

    2004  Volume 214, Issue 1-2, Page(s) 127–135

    Abstract: Dehydroepiandrosterone (DHEA) is synthesized in the brain, but whether DHEA is involved in modulating neuronal cell survival is not yet fully understood. Herein we show that when deprived of trophic support, GT1-7 hypothalamic neurons undergo apoptosis ... ...

    Abstract Dehydroepiandrosterone (DHEA) is synthesized in the brain, but whether DHEA is involved in modulating neuronal cell survival is not yet fully understood. Herein we show that when deprived of trophic support, GT1-7 hypothalamic neurons undergo apoptosis following exposure to DHEA, as demonstrated both by morphological and biochemical criteria. This proapoptotic effect appeared to be specific to DHEA itself, and not through conversion of DHEA to other steroids such as androgen or estrogen. Importantly, we determined that IGF-I protects GT1-7 neurons from DHEA-induced cell death. DHEA-induced apoptosis was associated with increased activation of caspase 3 and decreased PARP, which were both attenuated with addition of IGF-I. Addition of DHEA prevented phosphorylation of both Akt and glycogen synthase kinase-3 beta (GSK-3beta), downstream effector molecules of the phosphatidylinositol 3-kinase (PI3K) pathway. Further IGF-I was able to sustain Akt activity and thus preventing GSK-3beta activation in the presence of DHEA. On the other hand, the MAP kinases, ERK, p38, and JNK, were not affected by DHEA. These findings suggest that in GT1-7 hypothalamic neurons, DHEA acts detrimentally to induce cell death and IGF-I is able to rescue the neurons by preserving the activity of Akt, and therefore maintaining the proapoptotic kinase GSK-3beta, in a phosphorylated catalytically inactive state.
    MeSH term(s) Animals ; Apoptosis/drug effects ; Caspase 3 ; Caspases ; Cell Line, Transformed ; Dehydroepiandrosterone/pharmacology ; Glycogen Synthase Kinase 3/metabolism ; Glycogen Synthase Kinase 3 beta ; Hypothalamus/cytology ; Insulin-Like Growth Factor I/physiology ; Mice ; Neurons/cytology ; Neurons/metabolism ; Phosphorylation ; Protein-Serine-Threonine Kinases ; Proto-Oncogene Proteins/metabolism ; Proto-Oncogene Proteins c-akt ; Signal Transduction
    Chemical Substances Proto-Oncogene Proteins ; Dehydroepiandrosterone (459AG36T1B) ; Insulin-Like Growth Factor I (67763-96-6) ; Glycogen Synthase Kinase 3 beta (EC 2.7.11.1) ; Gsk3b protein, mouse (EC 2.7.11.1) ; Protein-Serine-Threonine Kinases (EC 2.7.11.1) ; Proto-Oncogene Proteins c-akt (EC 2.7.11.1) ; Glycogen Synthase Kinase 3 (EC 2.7.11.26) ; Casp3 protein, mouse (EC 3.4.22.-) ; Caspase 3 (EC 3.4.22.-) ; Caspases (EC 3.4.22.-)
    Language English
    Publishing date 2004-02-12
    Publishing country Ireland
    Document type Journal Article ; Research Support, Non-U.S. Gov't
    ZDB-ID 187438-x
    ISSN 1872-8057 ; 0303-7207
    ISSN (online) 1872-8057
    ISSN 0303-7207
    DOI 10.1016/j.mce.2003.10.064
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  6. Article ; Online: Diurnal profiling of neuroendocrine genes in murine heart, and shift in proopiomelanocortin gene expression with pressure-overload cardiac hypertrophy.

    Chalmers, Jennifer A / Lin, Shuo-Yen J / Martino, Tami A / Arab, Sara / Liu, Peter / Husain, Mansoor / Sole, Michael J / Belsham, Denise D

    Journal of molecular endocrinology

    2008  Volume 41, Issue 3, Page(s) 117–124

    Abstract: Neuroendocrine peptides express biologic activity relevant to the cardiovascular system, including regulating heart rate and blood pressure, though little is known about the mechanisms involved. Here, we investigated neuroendocrine gene expression ... ...

    Abstract Neuroendocrine peptides express biologic activity relevant to the cardiovascular system, including regulating heart rate and blood pressure, though little is known about the mechanisms involved. Here, we investigated neuroendocrine gene expression underlying diurnal physiology of the heart. We first used microarray and RT-PCR analysis and demonstrate the simultaneous expression of neuroendocrine genes in normal murine heart, including POMC, GnRH, neuropeptide Y, leptin receptor, GH-releasing hormone, cocaine- and amphetamine-regulated transcript, proglucagon, and galanin. We examined diurnal gene expression profiles, with cosinar bioinformatics to evaluate statistically significant rhythms. The POMC gene exhibits a day/night, circadian or diurnal, pattern of expression in heart, and we postulated that this may be important to cardiac growth and renewal. POMC diurnal gene rhythmicity is altered in pressure-overload cardiac hypertrophy, when compared with control heart, and levels increased at the dark-to-light transition times. These findings are also consistent with the proposal that neuropeptides mediate adverse remodeling processes, such as occur in pathologic hypertrophy. To investigate cellular responses, we screened three cell lines representing fibroblasts, cardiac myocytes, and vascular smooth muscle cells (NIH3T3, heart line 1, and mouse vascular smooth muscle cell line 1 (Movas-1) respectively). POMC mRNA expression is the most notable in Movas-1 cells and, furthermore, exhibits rhythmicity with culture synchronization. Taken together, these results highlight the diverse neuroendocrine mRNA expression profiles in cardiovasculature, and provide a novel model vascular culture system to research the role these neuropeptides play in organ health, integrity, and disease.
    MeSH term(s) Animals ; Aorta/pathology ; Blood Pressure ; Cardiomegaly/genetics ; Cardiomegaly/physiopathology ; Circadian Rhythm/physiology ; Constriction, Pathologic/genetics ; Gene Expression Profiling ; Gene Expression Regulation ; Male ; Mice ; Mice, Inbred C57BL ; Myocardium/metabolism ; Myocytes, Smooth Muscle/metabolism ; NIH 3T3 Cells ; Neuropeptides/genetics ; Neuropeptides/metabolism ; Neurosecretory Systems/metabolism ; Oligonucleotide Array Sequence Analysis ; Pro-Opiomelanocortin/genetics ; RNA, Messenger/genetics ; RNA, Messenger/metabolism ; Reverse Transcriptase Polymerase Chain Reaction ; Time Factors
    Chemical Substances Neuropeptides ; RNA, Messenger ; Pro-Opiomelanocortin (66796-54-1)
    Language English
    Publishing date 2008-09
    Publishing country England
    Document type Journal Article ; Research Support, Non-U.S. Gov't
    ZDB-ID 645012-x
    ISSN 1479-6813 ; 0952-5041
    ISSN (online) 1479-6813
    ISSN 0952-5041
    DOI 10.1677/JME-08-0050
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

To top