Reference Twice: A Simple and Unified Baseline for Few-Shot Instance Segmentation.

Few-Shot Instance Segmentation (FSIS) requires detecting and segmenting novel classes with limited support examples. Existing methods based on Region Proposal Networks (RPNs) face two issues: 1) Overfitting suppresses novel class objects; 2) Dual-branch models require complex spatial correlation strategies to prevent spatial information loss when generating class prototypes. We introduce a unified framework, Reference Twice (RefT), to exploit the relationship between support and query features for FSIS and related tasks. Our three main contributions are: 1) A novel transformer-based baseline that avoids overfitting, offering a new direction for FSIS; 2) Demonstrating that support object queries encode key factors after base training, allowing query features to be enhanced twice at both feature and query levels using simple cross-attention, thus avoiding complex spatial correlation interaction; 3) Introducing a class-enhanced base knowledge distillation loss to address the issue of DETR-like models struggling with incremental settings due to the input projection layer, enabling easy extension to incremental FSIS. Extensive experimental evaluations on the COCO dataset under three FSIS settings demonstrate that our method performs favorably against existing approaches across different shots, e.g., +8.2/ + 9.4 performance gain over state-of-the-art methods with 10/30-shots. Source code and models will be available at this github site.

Towards Open Vocabulary Learning: A Survey.

In the field of visual scene understanding, deep neural networks have made impressive advancements in various core tasks like segmentation, tracking, and detection. However, most approaches operate on the close-set assumption, meaning that the model can only identify pre-defined categories that are present in the training set. Recently, open vocabulary settings were proposed due to the rapid progress of vision language pre-training. These new approaches seek to locate and recognize categories beyond the annotated label space. The open vocabulary approach is more general, practical, and effective than weakly supervised and zero-shot settings. This paper thoroughly reviews open vocabulary learning, summarizing and analyzing recent developments in the field. In particular, we begin by juxtaposing open vocabulary learning with analogous concepts such as zero-shot learning, open-set recognition, and out-of-distribution detection. Subsequently, we examine several pertinent tasks within the realms of segmentation and detection, encompassing long-tail problems, few-shot, and zero-shot settings. As a foundation for our method survey, we first elucidate the fundamental principles of detection and segmentation in close-set scenarios. Next, we examine various contexts where open vocabulary learning is employed, pinpointing recurring design elements and central themes. This is followed by a comparative analysis of recent detection and segmentation methodologies in commonly used datasets and benchmarks. Our review culminates with a synthesis of insights, challenges, and discourse on prospective research trajectories. To our knowledge, this constitutes the inaugural exhaustive literature review on open vocabulary learning.

Omni-Frequency Channel-Selection Representations for Unsupervised Anomaly Detection.

Density-based and classification-based methods have ruled unsupervised anomaly detection in recent years, while reconstruction-based methods are rarely mentioned for the poor reconstruction ability and low performance. However, the latter requires no costly extra training samples for the unsupervised training that is more practical, so this paper focuses on improving reconstruction-based method and proposes a novel O mni-frequency C hannel-selection R econstruction (OCR-GAN) network to handle sensory anomaly detection task in a perspective of frequency. Concretely, we propose a Frequency Decoupling (FD) module to decouple the input image into different frequency components and model the reconstruction process as a combination of parallel omni-frequency image restorations, as we observe a significant difference in the frequency distribution of normal and abnormal images. Given the correlation among multiple frequencies, we further propose a Channel Selection (CS) module that performs frequency interaction among different encoders by adaptively selecting different channels. Abundant experiments demonstrate the effectiveness and superiority of our approach over different kinds of methods, e.g., achieving a new state-of-the-art 98.3 detection AUC on the MVTec AD dataset without extra training data that markedly surpasses the reconstruction-based baseline by +38.1 ↑ and the current SOTA method by +0.3 ↑ . The source code is available in the additional materials.

Pentapeptide-Zinc Chelate from Sweet Almond Expeller Amandin Hydrolysates: Structural and Physicochemical Characteristics, Stability and Zinc Transport Ability In Vitro.

To promote the application of almond expellers, sweet almond expeller globulin (amandin) was extracted for the preparation of bioactive peptides. After dual enzymatic hydrolysis, Sephadex G-15 gel isolation, reverse-phase high-performance liquid chromatography purification and ESI-MS/MS analysis, two novel peptides Val-Asp-Leu-Val-Ala-Glu-Val-Pro-Arg-Gly-Leu (1164.45 Da) and Leu-Asp-Arg-Leu-Glu (644.77 Da) were identified in sweet almond expeller amandin hydrolysates. Leu-Asp-Arg-Leu-Glu (LDRLE) of excellent zinc-chelating capacity (24.73 mg/g) was selected for preparation of peptide-zinc chelate. Structural analysis revealed that zinc ions were mainly bonded to amino group and carboxyl group of LDRLE. Potential toxicity and some physicochemical properties of LDRLE and Val-Asp-Leu-Val-Ala-Glu-Val-Pro-Arg-Gly-Leu (VDLVAEVPRGL) were predicted in silico. The results demonstrated that both LDRLE and VDLVAEVPRGL were not toxic. Additionally, zinc solubility of LDRLE-zinc chelate was much higher than that of zinc sulphate and zinc gluconate at pH 6.0−10.0 and against gastrointestinal digestion at 37 °C (p < 0.05). However, incubation at 100 °C for 20−60 min significantly reduced zinc-solubility of LDRLE-zinc chelate. Moreover, the chelate showed higher zinc transport ability in vitro than zinc sulphate and zinc gluconate (p < 0.05). Therefore, peptides isolated from sweet almond expeller amandin have potential applications as ingredient of zinc supplements.

Analysis of volatile organic compounds and metabolites of three cultivars of asparagus (Asparagus officinalis L.) using E-nose, GC-IMS, and LC-MS/MS.

Asparagus (A. officinalis L.) is a perennial herb of the genus Asparagus that is rich in nutrients. This study aimed to distinguish three cultivars of asparagus (Paladin, Grace, and Jinggang red) based on their volatile organic compounds (VOCs) and metabolic profiles. VOCs in the three cultivars were separated and identified using electronic nose (E-nose) and gas chromatography-ion mobility spectrometry (GC-IMS). Differences in metabolites among the three cultivars were analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). E-nose and GC-IMS showed that the VOCs in asparagus differed significantly among the three groups. E-nose result showed that purple asparagus (Jinggang red) was connected to a stronger earthy odor; green asparagus (Paladin and Grace) were shown characteristic dill flavor. Moreover, 50 VOCs were detected by using GC-IMS. Ketones and alcohols were most abundant in Paladin; methyl benzoate and dimethyl sulfide were most abundance in Grace; aldehydes and acids were most abundance in Jinggang red. Moreover, 130 and 71 different metabolites were detected in the positive and negative modes among three cultivars, such as quercetin and rutin. Functional analysis revealed that these metabolites were involved in beta-alanine metabolism and ATP-binding cassette (ABC) transporters. In summary, E-nose combined with GC-IMS and LC-MS/MS methods has good application prospects in evaluating and identifying VOCs and metabolites of different cultivars of asparagus. The identified VOCs and metabolites can provide guidelines for the development of functional asparagus products.

Adding Before Pruning: Sparse Filter Fusion for Deep Convolutional Neural Networks via Auxiliary Attention.

Filter pruning is a significant feature selection technique to shrink the existing feature fusion schemes (especially on convolution calculation and model size), which helps to develop more efficient feature fusion models while maintaining state-of-the-art performance. In addition, it reduces the storage and computation requirements of deep neural networks (DNNs) and accelerates the inference process dramatically. Existing methods mainly rely on manual constraints such as normalization to select the filters. A typical pipeline comprises two stages: first pruning the original neural network and then fine-tuning the pruned model. However, choosing a manual criterion can be somehow tricky and stochastic. Moreover, directly regularizing and modifying filters in the pipeline suffer from being sensitive to the choice of hyperparameters, thus making the pruning procedure less robust. To address these challenges, we propose to handle the filter pruning issue through one stage: using an attention-based architecture that adaptively fuses the filter selection with filter learning in a unified network. Specifically, we present a pruning method named adding before pruning (ABP) to make the model focus on the filters of higher significance by training instead of man-made criteria such as norm, rank, etc. First, we add an auxiliary attention layer into the original model and set the significance scores in this layer to be binary. Furthermore, to propagate the gradients in the auxiliary attention layer, we design a specific gradient estimator and prove its effectiveness for convergence in the graph flow through mathematical derivation. In the end, to relieve the dependence on the complicated prior knowledge for designing the thresholding criterion, we simultaneously prune and train the filters to automatically eliminate network redundancy with recoverability. Extensive experimental results on the two typical image classification benchmarks, CIFAR-10 and ILSVRC-2012, illustrate that the proposed approach performs favorably against previous state-of-the-art filter pruning algorithms.