LIVIVO - Search results -

Search results

Result 1 - 10 of total 32

Search options

Article ; Online: Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting.

Zhang, Kaidong / Peng, Jialun / Fu, Jingjing / Liu, Dong

IEEE transactions on pattern analysis and machine intelligence

2024 Volume PP

Abstract: Transformers have been widely used for video processing owing to the multi-head self attention (MHSA) mechanism. However, the MHSA mechanism encounters an intrinsic difficulty for video inpainting, since the features associated with the corrupted regions ...

Abstract	Transformers have been widely used for video processing owing to the multi-head self attention (MHSA) mechanism. However, the MHSA mechanism encounters an intrinsic difficulty for video inpainting, since the features associated with the corrupted regions are degraded and incur inaccurate self attention. This problem, termed query degradation, may be mitigated by first completing optical flows and then using the flows to guide the self attention, which was verified in our previous work - flow-guided transformer (FGT). We further exploit the flow guidance and propose FGT++ to pursue more effective and efficient video inpainting. First, we design a lightweight flow completion network by using local aggregation and edge loss. Second, to address the query degradation, we propose a flow guidance feature integration module, which uses the motion discrepancy to enhance the features, together with a flow-guided feature propagation module that warps the features according to the flows. Third, we decouple the transformer along the temporal and spatial dimensions, where flows are used to select the tokens through a temporally deformable MHSA mechanism, and global tokens are combined with the inner-window local tokens through a dual-perspective MHSA mechanism. FGT++ is experimentally evaluated to be outperforming the existing video inpainting networks qualitatively and quantitatively.
Language	English
Publishing date	2024-02-01
Publishing country	United States
Document type	Journal Article
ISSN	1939-3539
ISSN (online)	1939-3539
DOI	10.1109/TPAMI.2024.3361010
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

This service is chargeable due to the Delivery terms set by subito. Orders including an article and supplementary material will be classified as separate orders. In these cases, fees will be demanded for each order.

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Customized Segment Anything Model for Medical Image Segmentation

Zhang, Kaidong / Liu, Dong

2023

Abstract: We propose SAMed, a general solution for medical image segmentation. Different from the previous methods, SAMed is built upon the large-scale image segmentation model, Segment Anything Model (SAM), to explore the new research paradigm of customizing ... ...

Abstract	We propose SAMed, a general solution for medical image segmentation. Different from the previous methods, SAMed is built upon the large-scale image segmentation model, Segment Anything Model (SAM), to explore the new research paradigm of customizing large-scale models for medical image segmentation. SAMed applies the low-rank-based (LoRA) finetuning strategy to the SAM image encoder and finetunes it together with the prompt encoder and the mask decoder on labeled medical image segmentation datasets. We also observe the warmup finetuning strategy and the AdamW optimizer lead SAMed to successful convergence and lower loss. Different from SAM, SAMed could perform semantic segmentation on medical images. Our trained SAMed model achieves 81.88 DSC and 20.64 HD on the Synapse multi-organ segmentation dataset, which is on par with the state-of-the-art methods. We conduct extensive experiments to validate the effectiveness of our design. Since SAMed only updates a small fraction of the SAM parameters, its deployment cost and storage cost are quite marginal in practical usage. The code of SAMed is available at https://github.com/hitachinsk/SAMed. Comment: Technical report, 14 pages
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	004
Publishing date	2023-04-26
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting

Zhang, Kaidong / Peng, Jialun / Fu, Jingjing / Liu, Dong

2023

Abstract	Transformers have been widely used for video processing owing to the multi-head self attention (MHSA) mechanism. However, the MHSA mechanism encounters an intrinsic difficulty for video inpainting, since the features associated with the corrupted regions are degraded and incur inaccurate self attention. This problem, termed query degradation, may be mitigated by first completing optical flows and then using the flows to guide the self attention, which was verified in our previous work - flow-guided transformer (FGT). We further exploit the flow guidance and propose FGT++ to pursue more effective and efficient video inpainting. First, we design a lightweight flow completion network by using local aggregation and edge loss. Second, to address the query degradation, we propose a flow guidance feature integration module, which uses the motion discrepancy to enhance the features, together with a flow-guided feature propagation module that warps the features according to the flows. Third, we decouple the transformer along the temporal and spatial dimensions, where flows are used to select the tokens through a temporally deformable MHSA mechanism, and global tokens are combined with the inner-window local tokens through a dual perspective MHSA mechanism. FGT++ is experimentally evaluated to be outperforming the existing video inpainting networks qualitatively and quantitatively. Comment: This manuscript is a journal extension of our ECCV 2022 paper (arXiv:2208.06768)
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	004
Publishing date	2023-01-24
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: A Dataset for Deep Learning-based Bone Structure Analyses in Total Hip Arthroplasty

Zhang, Kaidong / Gan, Ziyang / Liu, Dong / Shang, Xifu

2023

Abstract: Total hip arthroplasty (THA) is a widely used surgical procedure in orthopedics. For THA, it is of clinical significance to analyze the bone structure from the CT images, especially to observe the structure of the acetabulum and femoral head, before the ... ...

Abstract	Total hip arthroplasty (THA) is a widely used surgical procedure in orthopedics. For THA, it is of clinical significance to analyze the bone structure from the CT images, especially to observe the structure of the acetabulum and femoral head, before the surgical procedure. For such bone structure analyses, deep learning technologies are promising but require high-quality labeled data for the learning, while the data labeling is costly. We address this issue and propose an efficient data annotation pipeline for producing a deep learning-oriented dataset. Our pipeline consists of non-learning-based bone extraction (BE) and acetabulum and femoral head segmentation (AFS) and active-learning-based annotation refinement (AAR). For BE we use the classic graph-cut algorithm. For AFS we propose an improved algorithm, including femoral head boundary localization using first-order and second-order gradient regularization, line-based non-maximum suppression, and anatomy prior-based femoral head extraction. For AAR, we refine the algorithm-produced pseudo labels with the help of trained deep models: we measure the uncertainty based on the disagreement between the original pseudo labels and the deep model predictions, and then find out the samples with the largest uncertainty to ask for manual labeling. Using the proposed pipeline, we construct a large-scale bone structure analyses dataset from more than 300 clinical and diverse CT scans. We perform careful manual labeling for the test set of our data. We then benchmark multiple state-of-the art deep learning-based methods of medical image segmentation using the training and test sets of our data. The extensive experimental results validate the efficacy of the proposed data annotation pipeline. The dataset, related codes and models will be publicly available at https://github.com/hitachinsk/THA. Comment: 16 pages, 17 figures
Keywords	Electrical Engineering and Systems Science - Image and Video Processing ; Computer Science - Computer Vision and Pattern Recognition
Subject code	006
Publishing date	2023-06-07
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article: [Two times of acute symptomatic epidural hematoma after lumbar posterior internal fixation:a case report].

Yang, Jun / Deng, Qiang / Qiao, Xiao-Wan / Zhu, Bao / Zhang, Kai-Dong / Yang, Hai-Yun

Zhongguo gu shang = China journal of orthopaedics and traumatology

2023 Volume 36, Issue 9, Page(s) 803–808

MeSH term(s)	Humans ; Hematoma, Epidural, Spinal
Language	Chinese
Publishing date	2023-09-21
Publishing country	China
Document type	Case Reports ; Journal Article
ISSN	1003-0034
ISSN	1003-0034
DOI	10.12200/j.issn.1003-0034.2023.09.003
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Towards Interactive Image Inpainting via Sketch Refinement

Liu, Chang / Xu, Shunxin / Peng, Jialun / Zhang, Kaidong / Liu, Dong

2023

Abstract: One tough problem of image inpainting is to restore complex structures in the corrupted regions. It motivates interactive image inpainting which leverages additional hints, e.g., sketches, to assist the inpainting process. Sketch is simple and intuitive ... ...

Abstract	One tough problem of image inpainting is to restore complex structures in the corrupted regions. It motivates interactive image inpainting which leverages additional hints, e.g., sketches, to assist the inpainting process. Sketch is simple and intuitive to end users, but meanwhile has free forms with much randomness. Such randomness may confuse the inpainting models, and incur severe artifacts in completed images. To address this problem, we propose a two-stage image inpainting method termed SketchRefiner. In the first stage, we propose using a cross-correlation loss function to robustly calibrate and refine the user-provided sketches in a coarse-to-fine fashion. In the second stage, we learn to extract informative features from the abstracted sketches in the feature space and modulate the inpainting process. We also propose an algorithm to simulate real sketches automatically and build a test protocol with different applications. Experimental results on public datasets demonstrate that SketchRefiner effectively utilizes sketch information and eliminates the artifacts due to the free-form sketches. Our method consistently outperforms the state-of-the-art ones both qualitatively and quantitatively, meanwhile revealing great potential in real-world applications. Our code and dataset are available. Comment: Fix some errors, polish the paper
Keywords	Computer Science - Computer Vision and Pattern Recognition
Subject code	004
Publishing date	2023-06-01
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Flow-Guided Transformer for Video Inpainting

Zhang, Kaidong / Fu, Jingjing / Liu, Dong

2022

Abstract: We propose a flow-guided transformer, which innovatively leverage the motion discrepancy exposed by optical flows to instruct the attention retrieval in transformer for high fidelity video inpainting. More specially, we design a novel flow completion ... ...

Abstract	We propose a flow-guided transformer, which innovatively leverage the motion discrepancy exposed by optical flows to instruct the attention retrieval in transformer for high fidelity video inpainting. More specially, we design a novel flow completion network to complete the corrupted flows by exploiting the relevant flow features in a local temporal window. With the completed flows, we propagate the content across video frames, and adopt the flow-guided transformer to synthesize the rest corrupted regions. We decouple transformers along temporal and spatial dimension, so that we can easily integrate the locally relevant completed flows to instruct spatial attention only. Furthermore, we design a flow-reweight module to precisely control the impact of completed flows on each spatial transformer. For the sake of efficiency, we introduce window partition strategy to both spatial and temporal transformers. Especially in spatial transformer, we design a dual perspective spatial MHSA, which integrates the global tokens to the window-based attention. Extensive experiments demonstrate the effectiveness of the proposed method qualitatively and quantitatively. Codes are available at https://github.com/hitachinsk/FGT. Comment: ECCV 2022
Keywords	Computer Science - Computer Vision and Pattern Recognition
Publishing date	2022-08-13
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article ; Online: First Report of Phytophthora Blight Caused by Phytophthora nicotianae on Daphne odora in China

Hu, Jiangtao / Zhou, Ying / Luo, Sumei / Zhang, Yuanfu / Chen, Yuanhua / Cai, Lei / Zhou, Yonghui / Li, Rong / Zhang, Kaidong / Liu, Shuyuan / Liu, Xiaoping

Plant Disease. 2023 June 01, v. 107, no. 6 p.1953-

2023

Abstract: The variegated leaves and fragrant flowers of Daphne odora var. marginata Mak. make it a popular garden plant. In May 2020, we found diseased D. odora plants in a greenhouse at the Ganzhou Vegetable and Flower Research Institute in southeast China. ... ...

Abstract	The variegated leaves and fragrant flowers of Daphne odora var. marginata Mak. make it a popular garden plant. In May 2020, we found diseased D. odora plants in a greenhouse at the Ganzhou Vegetable and Flower Research Institute in southeast China. Seventy-two percent of 1,800 plants had Phytophthora blight-like symptoms, including shrunken stems; black withered branches; wilted and dropped leaves; and rotted and dark green roots. The root and stem tissue surfaces were disinfected with 75% ethanol for 30 s followed by 0.1% HgCl₂ for 1 min, rinsed thrice with sterile water, and cultured on potato-dextrose agar (PDA) medium at 25°C. Mycelia from the diseased tissue were subcultured on fresh PDA medium, providing three colonies. White colonies (∼4.1 mm) were formed after 10 days at 25°C. Sporangia and chlamydospores were induced by placing actively growing mycelia on PDA medium at 25°C for ∼30 days and then at 45°C for ∼3 days. Sporangia were ovoid to spherical and 19.33 × 20.99 μm, whereas chlamydospores were spherical and 15.68 × 16.10 μm. All three colonies resembled Phytophthora spp. Genomic DNA was extracted from isolates using the Ezup Column Fungi Genomic DNA Purification Kit (Sangon Biotech, Shanghai), and rDNA-ITS and β-tubulin were amplified and sequenced. BLAST analysis (GenBank) revealed that the ITS (accession no. MZ676071) and β-tubulin (MZ748503) sequences of isolates shared the highest similarity (99 to 100%) with those of Phytophthora nicotianae (Duccio et al. 2015). A phylogenetic tree of the relationship between our isolate hjt3 and its close relatives within the P. nicotianae species was constructed using the MEGA X neighbor-joining method. The pathogen was identified as P. nicotianae based on morphological and molecular characteristics. Sequencing results of the three samples were consistent, all indicating P. nicotianae. A specimen (JXAU-H2020245) was deposited in the Herbarium of the College of Agronomy, Jiangxi Agricultural University. To confirm pathogenicity, 9-month-old, healthy D. odora plants were used for stem and soil inoculation. Stems were cut ∼5 cm from the soil with sterilized scalpels and inoculated with 0.8-cm diameter PDA plugs containing actively growing mycelia of isolate hjt3. The soil was sterilized and 0.8-cm PDA plugs containing actively growing mycelia were buried in the soil at ∼5 cm; the mycelia were in contact with the roots. Plants in both groups were treated equally; those inoculated with sterile PDA plugs served as controls. There were six plants in each group, with each experiment performed in triplicate. All plants were incubated in a greenhouse at 25 to 28°C. The stems shrank and began to rot rapidly after 7 days and the branches turned black and withered within 2 weeks. After soil inoculation, the stems of the inoculated plants blackened and rotted in ∼20 days and the roots rotted and turned dark green. These symptoms rapidly spread to the branches. The control plants did not exhibit any symptoms. Reisolated colonies showed the same morphological traits as the isolates used forinoculation; no target colonies were isolated from the control plants.Phytophthora blight caused by P. nicotianae on D. odora has been reported in Italy (Garibaldi et al. 2009) and Korea (Kwon et al. 2005). This is the first detection in China. Therefore, Phytophthora blight on D. odora caused by P. nicotianae should be monitored and controlled to promote the development of the D. odora industry.
Keywords	DNA ; Daphne odora ; Phytophthora nicotianae ; agar ; agricultural colleges ; agronomy ; blight ; chlamydospores ; ethanol ; flowers ; greenhouses ; herbaria ; industry ; mycelium ; ornamental plants ; pathogenicity ; pathogens ; phylogeny ; research institutions ; soil ; soil inoculation ; sporangia ; vegetables ; China ; Italy ; Korean Peninsula ; herbaceous/flowering plants ; ornamentals ; pathogen detection ; tropical plants
Language	English
Dates of publication	2023-0601
Publishing place	The American Phytopathological Society
Document type	Article ; Online
ZDB-ID	754182-x
ISSN	0191-2917
ISSN	0191-2917
DOI	10.1094/PDIS-08-22-1994-PDN
Database	NAL-Catalogue (AGRICOLA)

In stock of ZB MED Bonn / Germany

Z 2043: Show issues

Order via subito

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Details ▾
- See ZB MED holdings
- Order with fees

Article: Post-traumatic cauda equina nerve calcification: A case report.

Liu, Yan-Dong / Deng, Qiang / Li, Jun-Jie / Yang, Hai-Yun / Han, Xian-Fu / Zhang, Kai-Dong / Peng, Ran-Dong / Xiang, Qian-Qian

World journal of clinical cases

2022 Volume 11, Issue 6, Page(s) 1356–1364

Abstract: Background: Post-traumatic cauda equina nerve calcification is extremely rare in clinical practice, and its etiology, pathogenesis, treatment and prognosis are unclear. There are few studies and reports on Post-traumatic cauda equina nerve calcification, ...

Abstract	Background: Post-traumatic cauda equina nerve calcification is extremely rare in clinical practice, and its etiology, pathogenesis, treatment and prognosis are unclear. There are few studies and reports on Post-traumatic cauda equina nerve calcification, and this review reports a case of Post-traumatic cauda equina nerve calcification for reference. Case summary: A 52-year-old patient presented to our hospital with a history of lumbar spinal stenosis and a lumbar vertebral fracture caused by trauma. The patient's right lower limb had weakness in hip flexion, knee extension and plantarflexion with muscle strength grade 3, right ankle dorsiflexion and thumb dorsiflexion with muscle strength grade 0. The patient's skin sensation below the right knee plane disappeared. The patient's Computed tomography (CT) data showed signs of cauda equina nerve calcification and the terminal filaments in the plane of the third to fifth lumbar vertebrae. After treatment the patient's symptoms were slightly relieved. Conclusion: We provide an extremely rare case of Post-traumatic cauda equina nerve calcification and offer a conservative treatment plan. However, the etiology, mechanism and treatment of Post-traumatic cauda equina nerve calcification are still unclear. This requires scholars to conduct more research and exploration in this area.
Language	English
Publishing date	2022-04-19
Publishing country	United States
Document type	Case Reports
ISSN	2307-8960
ISSN	2307-8960
DOI	10.12998/wjcc.v11.i6.1356
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: DamWorld

Ren, Pengzhen / Zhang, Kaidong / Zheng, Hetao / Li, Zixuan / Wen, Yuhang / Zhu, Fengda / Ma, Mas / Liang, Xiaodan

Progressive Reasoning with World Models for Robotic Manipulation

2023

Abstract: The research on embodied AI has greatly promoted the development of robot manipulation. However, it still faces significant challenges in various aspects such as benchmark construction, multi-modal perception and decision-making, and physical execution. ... ...

Abstract	The research on embodied AI has greatly promoted the development of robot manipulation. However, it still faces significant challenges in various aspects such as benchmark construction, multi-modal perception and decision-making, and physical execution. Previous robot manipulation simulators were primarily designed to enrich manipulation types and types of objects while neglecting the balance between physical manipulation and language instruction complexity in multi-modal environments. This paper proposes a new robot manipulation simulator and builds a comprehensive and systematic robot manipulation benchmark with progressive reasoning tasks called SeaWave (i.e., a progressive reasoning benchmark). It provides a standard test platform for embedded AI agents in a multi-modal environment, which can evaluate and execute four levels of human natural language instructions at the same time. Previous world model-based robot manipulation work lacked research on the perception and decision-making of complex instructions in multi-modal environments. To this end, we propose a new world model tailored for cross-modal robot manipulation called DamWorld. Specifically, DamWorld takes the current visual scene and predicted execution actions based on natural language instructions as input, and uses the next action frame to supervise the output of the world model to force the model to learn robot manipulation consistent with world knowledge. Compared with the renowned baselines (e.g., RT-1), our DamWorld improves the manipulation success rate by 5.6% on average on four levels of progressive reasoning tasks. It is worth noting that on the most challenging level 4 manipulation task, DamWorld still improved by 9.0% compared to prior works.
Keywords	Computer Science - Robotics ; Computer Science - Artificial Intelligence ; Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Machine Learning
Subject code	629
Publishing date	2023-06-20
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

To top

More links

Kategorien

Order via subito

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

More links

Kategorien

In stock of ZB MED Bonn / Germany

Order via subito

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED