The term out-of-stock (OOS) describes a problem that occurs when shoppers come to a store and the product they are seeking is not present on its designated shelf. Missing products generate huge sales losses and may lead to a declining reputation or the loss of loyal customers. In this paper, we propose a novel deep-learning (DL)-based OOS-detection method that utilizes a two-stage training process and a post-processing technique designed for the removal of inaccurate detections. To develop the method, we utilized an OOS detection dataset that contains a commonly used fully empty OOS class and a novel class that represents the frontal OOS. We present a new image augmentation procedure in which some existing OOS instances are enlarged by duplicating and mirroring themselves over nearby products. An object-detection model is first pre-trained using only augmented shelf images and, then, fine-tuned on the original data. During the inference, the detected OOS instances are post-processed based on their aspect ratio. In particular, the detected instances are discarded if their aspect ratio is higher than the maximum or lower than the minimum instance aspect ratio found in the dataset. The experimental results showed that the proposed method outperforms the existing DL-based OOS-detection methods and detects fully empty and frontal OOS instances with 86.3% and 83.7% of the average precision, respectively.

Hand gesture recognition is a rapidly expanding field with diverse applications, and the use of skeleton-based methods is gaining popularity due to their potential for lightweight execution on embedded devices. However, ensuring robustness and accuracy in both gesture classification and temporal localization is critical for any gesture recognition system to be successful. In this paper, we propose a novel skeleton-based approach to online gesture recognition that draws inspiration from the YOLO object detection model. Specifically, we propose a transformer-based architecture for online gesture recognition that directly predicts both gesture classes and gesture boundaries from a sliding window input. The model is trained in an end-to-end manner using a proposed loss function that focuses on samples containing gesture centers and learns to predict boundaries in the temporal domain analogous to object centers and spatial bounding boxes in YOLO object detection. To evaluate the effectiveness of our method, we conduct experiments on two publicly available continuous hand gesture datasets: SHREC’22 and IPN Hand. Our results outperform the results of skeleton-based approaches from the SHREC’22 online gesture recognition contest. Moreover, our approach obtained competitive results compared to the approaches utilizing alternative input modalities on the IPN Hand dataset.

The authors introduce a system for person de‐identification in video data that de‐identifies biometric and non‐biometric features, namely faces, hairstyles and clothing colours. The authors’ system detects human faces and silhouettes in the input video and replaces the detected faces with random synthesised faces obtained using a deep convolutional generative adversarial network. Alternative hairstyles are rendered over the synthesised faces, and the human silhouette is recoloured so that skin hues are preserved and clothing hues are altered. Through the use of artificially synthesised faces that look realistic, they ensure that the de‐identified image looks natural and at the same time avoid ethical and legal considerations present when using real face images as replacement faces. As they address non‐biometric feature de‐identification, their system offers a considerably higher level of privacy protection than commonly employed solutions that use simple image processing techniques such as blurring. Qualitative and quantitative evaluation suggests that their system produces de‐identified images that look natural, at the same time being resistant to re‐identification attacks.

Standard machine learning is unable to accommodate inputs which do not belong to the training distribution. The resulting models often give rise to confident incorrect predictions which may lead to devastating consequences. This problem is especially demanding in the context of dense prediction since input images may be only partially anomalous. Previous work has addressed dense out-of-distribution detection by discriminative training with respect to off-the-shelf negative datasets. However, real negative data may lead to over-optimistic evaluation due to possible overlap with test anomalies. To this end, we extend this approach by generating synthetic negative patches along the border of the inlier manifold. We leverage a jointly trained normalizing flow due to a coverage-oriented learning objective and the capability to generate samples at different resolutions. We detect anomalies according to a principled information-theoretic criterion which can be consistently applied through training and inference. The resulting models set the new state of the art on benchmarks for out-of-distribution detection in road-driving scenes and remote sensing imagery despite minimal computational overhead.

Product recognition is a fine-grained image retrieval problem because grocery stores can contain several thousand products on their shelves, some of which have minimal visual differences. Given that the products on the store’s shelves change frequently, it would be practical to build a system using a single reference image or a few per product. This task is very challenging in computer vision, whereas humans solve it more effortlessly by relying on contextual information. Our work incorporates semantic and spatial context into a novel product recognition method. We take advantage of stores being organized into different sections and aisles, where similar products are placed nearby. First, we propose the Hierarchical Auxiliary Loss (HAL) for learning an organized feature space in which products from the same category, usually placed in the same store section, are close to each other. Second, we propose the Context-Aware Query Expansion (CAQE) module for the inference phase, in which each feature vector of a product on a shelf is expanded with the feature vectors of neighboring products. The amount of information exchanged between the two products depends on the similarity of their feature vectors and spatial distance on the store shelves. To demonstrate the effectiveness of our contributions, we conducted detailed experiments on publicly available grocery product datasets and showed that our method achieves state-of-the-art results.