Recently, dropout has seen increasing use in deep learning. For deep convolutional neural networks, dropout is known to work well in fully-connected layers. However, its effect in convolutional and ...pooling layers is still not clear. This paper demonstrates that max-pooling dropout is equivalent to randomly picking activation based on a multinomial distribution at training time. In light of this insight, we advocate employing our proposed probabilistic weighted pooling, instead of commonly used max-pooling, to act as model averaging at test time. Empirical evidence validates the superiority of probabilistic weighted pooling. We also empirically show that the effect of convolutional dropout is not trivial, despite the dramatically reduced possibility of over-fitting due to the convolutional architecture. Elaborately designing dropout training simultaneously in max-pooling and fully-connected layers, we achieve state-of-the-art performance on MNIST, and very competitive results on CIFAR-10 and CIFAR-100, relative to other approaches without data augmentation. Finally, we compare max-pooling dropout and stochastic pooling, both of which introduce stochasticity based on multinomial distributions at pooling stage.
The encoder-decoder framework has proliferated in current image captioning task, where the decoder generates target description word by word based on the preceding captions. However, this framework ...encounters two main concerns. Firstly, the decoder cannot adequately capture global dependencies between the current predicted target word and all the previously generated words. Secondly, some generated words (e.g., “on”, “the” and “of”) provide insufficient information, which may deviate from the sentence semantics during gradually generating captions. To address above concerns, in this paper we propose a novel adaptive incremental global context attention (IGCA) method to capture the global information between target words, thus enhancing target word predictions in image captioning. Specifically, all of previous historical decoder hidden states are utilized as the global feature to guide the generation of subsequent word. During the generation procedure, the proposed IGCA mechanism is able to dynamically focus on these text features that are most correlated with the currently generated word. To verify the efficiency of our IGCA model, we conducted extensive experiments on the three public benchmark datasets. The experimental results demonstrate that the proposed model brings significant improvement over the conventional attention-based encoder-decoder methods and achieves state-of-the-art performance on Flick 30k and Flick 8k datasets.
The lymphatic network remodeling may guide tumor metastasis in a sentinel lymph node (SLN). Although tumor‐derived exosomes have been demonstrated to modify the microenvironment in adjacent organs ...and initiate a premetastatic niche, their influence on the lymphatic network in SLNs has not been explained. Here, we show that CT26 cell exosomes (Exo) promote the proliferation of lymphatic endothelial cells and the formation of lymphatic network in SLN, facilitating the SLN metastasis of colorectal cancer (CRC). Uptake of Exo by macrophages promoted VEGFC secretion both in vivo and in vitro. Exo increased the frequency of F4/80+ macrophages in the SLN. Macrophage ablation by clodrosome prevented the exosomal effect on lymphatic network remodeling and SLN metastasis. Exosomal IRF‐2 was highly expressed in serum exosomes isolated from CRC patients with LN metastasis relative to patients without LN metastasis and healthy controls. Mechanistically, exosomal IRF‐2 induced the release of VEGFC by macrophages. An IRF‐2 knockdown attenuated the lymphatic network remodeling in the SLN and suppressed the SLN metastasis. Our data suggest that exosomal IRF‐2 remodels the lymphatic network in an SLN and may predict the development of CRC LN metastases.
What's new?
Tumor‐derived exosomes have raised wide interest as intercellular messengers of oncogenic information serving to prepare a pre‐metastatic niche. The effects of cancer‐derived exosomes on lymphatic network remodeling in sentinel lymph nodes (SLNs) remain unclear, however. Here, the authors demonstrate that colorectal cancer‐derived exosomes promote lymphangiogenesis in SLNs through IRF‐2, which is capable of inducing VEGFC expression in macrophages. This may create a favorable environment for lymph node metastasis. The findings contribute to the understanding of the mechanisms underlying exosomal regulation of the tumor microenvironment and open up new ways to control tumor metastasis.
Grounding natural language in video aims at retrieving a matching moment in a long, untrimmed video described by a referring natural language query. It is a challenging issue due to the dominating ...influence from noise background in untrimmed video and the complex temporal relationships introduced by the query. Existing methods treat different candidate segments separately in a matching and aligning manner and thus neglect that different target segments require different levels of context information. In this paper, we present the semantic modulation residual module, a novel single-shot feed-forward residual network that explicitly integrates various temporal scale features and introduces less noise to the final moments representation with the guide of query semantic information. To establish more fine-grained interactions between different moments, a global interaction module is embedded in the network. Moreover, the data imbalance issue caused by the sparse annotated moments weakens the effect of binary cross-entropy criterion. Therefore, we design a foreground recalibration mechanism to enhance the intra-class consistency and highlight the positive moments. We evaluate our method on three benchmark datasets i.e., TACoS, Charades-STA and ActivityNet Captions, achieving state-of-the-art performance without any post-processing. In particular, we reach 32.17%, 45.11% and 43.76% under the metric Rank@1, IoU@0.5 on TACoS, Charades-STA and ActivityNet Captions, respectively. Furthermore, ablation studies were performed to show the effectiveness of individual components in our proposed method. We hope that the proposed method can serve as a strong and simple alternative for fine-grained video retrieval.
Convolutional Neural Networks (CNNs) have played an important role in saliency detection. How to detect a salient object as a whole is a key issue. However, most existing learning-based methods are ...not accurate enough to detect salient objects in complex scenes, such as easily overlooked small salient areas in a whole salient object, which is called scale imbalance problem in this paper. To address this issue, Scale Balance Network (SBN) based on fully convolutional network is proposed to accurately recognize and comprehensively detect salient objects. Firstly, to detect more small salient areas, a specially designed backbone instead of common backbone is adopted in this paper, which can capture larger resolution with more spatial features in deeper layers. Secondly, we present a novel progressive pyramid mechanism named Connective Feature Pyramid Module (CFPM), aiming to make the network focus on the balance between the large salient areas and the small ones. Finally, we present an Edge Enhancement Architecture with Various Kernels (EEAVK) to locate the saliency maps and refine the boundary features. Experimental results on five benchmark datasets show that the proposed SBN method achieves consistently superior performance in comparison with other state-of-the-art ones under different evaluation metrics.
We demonstrated a real-time scanning structured-light depth sensing system based on a solid-state vertical cavity surface-emitting laser (VCSEL) beam scanner integrated with an electro-thermally ...tunable VCSEL. Through a swept voltage added to the tunable VCSEL, a field of view of 6°×12° could be scanned with a scanning speed of 100 kHz by the beam scanner. Adopting the beam scanner, the real-time depth image with a lateral resolution of 10,000 (20×500) was obtained by measuring a step target placed at 35cm. The frame rate could be >10Hz even if sunlight shot noise is artificially added to the experimental data. By using a higher-speed camera, a potential lateral resolution could be reached at 50,000 (100×500) with a frame rate of > 20Hz. By using flat optics, a compact scanning module offering line pattern with FoV of >40°×20° was also demonstrated. It could help to realize high-resolution and high-accuracy structured-light sensing with a compact module.
Exosomal proteins are emerging as relevant diagnostic and prognostic biomarkers for cancer. This study was aimed at illustrating the clinical significance of exosomal Copine III (CPNE3) purified from ...the plasma of colorectal cancer (CRC) patients. The CPNE3 expression levels in CRC tissues were analyzed by real‐time PCR, western blot, and immunohistochemistry. Plasma exosomes were isolated to examine the CPNE3 level using ELISA. Pearson’s correlation analysis was performed to investigate the CPNE3 levels between CRC tissues and matched plasma samples. Receiver operating characteristic curve analysis was developed to measure the diagnostic performance of exosomal CPNE3. The Kaplan–Meier method and Cox's proportional hazards model were utilized to determine statistical differences in survival times. CPNE3 showed increased expressions in the CRC tissues. A moderately significant correlation was found between CPNE3 expression in CRC tissues by immunohistochemistry and matched serum exosomal CPNE3 expression by ELISA (r = 0.645,(r = 0.645, p < 0.001). < 0.001). Exosomal CPNE3 yielded a sensitivity of 67.5% and a specificity of 84.4% in CRC at the cutoff value of 0.143 pg per 1ug1 ug exosome. Combined data from carcinoembryonic antigen and exosomal CPNE3 achieved 84.8% sensitivity and 81.2% specificity as a diagnostic tool. CRC patients with lower exosomal CPNE3 levels had substantially better disease‐free survival (hazard ratio HR, 2.9; 95% confidence interval CI: 1.3–6.4; p = 0.009) = 0.009) and overall survival (HR, 3.4; 95% CI: 1.2–9.9; p = 0.026) = 0.026) compared with those with higher exosomal CPNE3 levels. Exosomal CPNE3 show potential implications in CRC diagnosis and prognosis.
Exosomal Copine III (CPNE3) levels are associated with the tumor extent and may serve as a diagnostic biomarker in patients with colorectal cancer (CRC). Combining measurement of exosomal CPNE3 in plasma with carcinoembryonic antigen (CEA) may further improve the diagnostic accuracy. The high exosomal CPNE3 level was a reliable indicator for worse disease‐free survival (DFS) and overall survival (OS).
Few shot class incremental learning (FSCIL) is an extremely challenging but valuable problem in real-world applications. When faced with novel few shot tasks in each incremental stage, it should take ...into account both catastrophic forgetting of old knowledge and overfitting of new categories with limited training data. In this paper, we propose an efficient prototype replay and calibration (EPRC) method with three stages to improve classification performance. We first perform effective pre-training with rotation and mix-up augmentations in order to obtain a strong backbone. Then a series of pseudo few shot tasks are sampled to perform meta-training, which enhances the generalization ability of both the feature extractor and projection layer and then helps mitigate the over-fitting problem of few shot learning. Furthermore, an even nonlinear transformation function is incorporated into the similarity computation to implicitly calibrate the generated prototypes of different categories and alleviate correlations among them. Finally, we replay the stored prototypes to relieve catastrophic forgetting and rectify prototypes to be more discriminative in the incremental-training stage via an explicit regularization within the loss function. The experimental results on CIFAR-100 and
ImageNet demonstrate that our EPRC significantly boosts the classification performance compared with existing mainstream FSCIL methods.
Dictionary learning plays an important role in sparse representation based face recognition. Many dictionary learning algorithms have been successfully applied to face recognition. However, for ...corrupted data because of noise or face variations (e.g. occlusion and large pose variation), their performances decline due to the disparity between domains. In this paper, we propose a face recognition algorithm based on dictionary learning and subspace learning (DLSL). In DLSL, a new subspace learning algorithm (SL) is proposed by using sparse constraint, low-rank technology and our label relaxation model to reduce the disparity between domains. Meanwhile, we propose a high-performance dictionary learning algorithm (HPDL) by constructing the embedding term, non-local self-similarity term, and time complexity drop term. In the obtained subspace, we use HPDL to classify these mapped test samples. DLSL is compared with other 28 algorithms on FRGC, LFW, CVL, Yale B and AR face databases. Experimental results show that DLSL achieves better performance than those 28 algorithms, including many state-of-the-art algorithms, such as recurrent regression neural network (RRNN), multimodal deep face recognition (MDFR) and projective low-rank representation (PLR).
PDF
