We are motivated by the observation that for problems where inputs and outputs are in the same form such as in image enhancement, deep neural networks can be reinforced by retraining the network using a new target set to the output for the original target. As an example, we introduce a new learning strategy for super-resolution by recurrently training the same simple network. Unlike the existing self-trained SR, which involves a single stage of learning with multiple runs at test time, our method trains the same SR network multiple times with increasingly better targets requiring only a single inference at test time. At each stage of the proposed learning scheme, a new target for training is obtained by applying the most recently trained SR network to the original image and downscaling the resultant SR image to normalize the size. Even if downscaling is involved, we argue that the downscaled SR image acts as a better target compared to the old target. We could mathematically demonstrate that this process is similar to unsharp masking when it is linearly approximated and that this process makes the image sharper. However, unlike unsharp masking, the proposed recurrent learning tends to converge to a specific target. By retraining the existing network aiming at a more enhanced target, the proposed method can achieve a similar effect of applying SR multiple times without increasing implementation cost and inference time. To objectively verify the supremacy of our approach by experiments, we propose to use VIQET MOS, which does not require a reference image as a measure of image quality. As far as we know, our work of using an objective quality measure in image enhancement is the first one whose validity was verified by showing similar results to the actual user's subjective evaluation. The proposed recurrent learning scheme makes existing SR algorithms more useful by clearly improving the effect of SR.
Video frame interpolation is the task of creating an interframe between two adjacent frames along the time axis. So, instead of simply averaging two adjacent frames to create an intermediate image, this operation should maintain semantic continuity with the adjacent frames. Most conventional methods use optical flow, and various tools such as occlusion handling and object smoothing are indispensable. Since the use of these various tools leads to complex problems, we tried to tackle the video interframe generation problem without using problematic optical flow. To enable this, we have tried to use a deep neural network with an invertible structure, and developed an invertible U-Net which is a modified normalizing flow. In addition, we propose a learning method with a new consistency loss in the latent space to maintain semantic temporal consistency between frames. The resolution of the generated image is guaranteed to be identical to that of the original images by using an invertible network. Furthermore, as it is not a random image like the ones by generative models, our network guarantees stable outputs without flicker. Through experiments, we confirmed the feasibility of the proposed algorithm and would like to suggest invertible U-Net as a new possibility for baseline in video frame interpolation. This paper is meaningful in that it is the world's first attempt to use invertible networks instead of optical flows for video interpolation.
Tumor-associated macrophages (TAMs) have been implicated in suppressing T-cell activity against cancer cells, representing an important target for the development of new immune therapy combinations. Accumulating evidence indicates that increased numbers of TAMs infiltrating cancer including melanoma are associated with poor prognosis. However, TAMs remain poorly defined in human cancers as they can exhibit a range of different molecular phenotypes and functional properties, and their cellular precursors are also uncertain, necessitating further investigation to enable their therapeutic manipulation. Using multiplex immunohistochemistry up to 7 colors and flow cytometry, here we sought to examine molecular characteristics of TAMs in human melanoma metastases to gain better understanding of the mechanisms driving their immune-suppressive activity in tumors as wells as the pathways recruiting their precursors. In lymph node and dermal melanoma metastases, we found that a population of TAMs expressing the marker CD163 frequently express molecules that are highly suppressive of T-cell function, including the PD-1 ligands PD-L1 and PD-L2, IDO (indoleamine 2,3-dioxygenase), TGF-β, and IL-10. In particular, the predominant source of PD-L1 is not the melanoma cells present but the infiltrating CD163+ TAMs. We also observed that many of the CD163+ TAMs downmodulate MHC class II molecules, represented by HLA-DR, suggesting they have suppressed capacity to present melanoma cell antigens to T cells. CD163+ TAMs in melanoma metastases often express weak levels of CD14 and CCR2, while cells adjacent to them express these markers brightly, suggesting that these TAMs are likely to derive from recently extravasated monocytes. Further analysis confirmed that CCL2, the chemokine ligand for CCR2, is indeed present in many melanoma tissue samples examined. Interestingly, the majority of CD163+CD14+ TAMs in melanoma metastases also express high levels of CD16, suggesting that these TAMs are either more closely related to the nonclassical CD14+CD16+ monocytes or derive from the classical CD14+CD16- monocytes that subsequently gain CD16 expression after entering the tumor tissue. In vitro, we were able to generate cells from primary human monocytes that recapitulate the phenotype of CD163+ TAMs observed in situ and are capable of suppressing T cells, which will provide us a valuable tool to carry out functional studies to reveal how TAMs suppress T-cell function and how to reverse such activities. Collectively, our data demonstrate the complex molecular features of CD163+ TAMs in melanoma metastases, necessitating further investigations to determine the dominant T cell-suppressive mechanisms used by TAMs and to prioritize therapeutic targets. Citation Format: Saem Park, Anna Brooks, Chun-Jen Chen, Rod Dunbar. Molecular characteristics of tumor-associated macrophages in human melanoma metastases [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2019 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2020;8(3 Suppl):Abstract nr B101.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.