Understanding the Limitations of Diffusion Concept Algebra Through Food

Image generation techniques, particularly latent diffusion models, have exploded in popularity in recent years. Many techniques have been developed to manipulate and clarify the semantic concepts these large-scale models learn, offering crucial insights into biases and concept relationships. However, these techniques are often only validated in conventional realms of human or animal faces and artistic style transitions. The food domain offers unique challenges through complex compositions and regional biases, which can shed light on the limitations and opportunities within existing methods. Through the lens of food imagery, we analyze both qualitative and quantitative patterns within a concept traversal technique. We reveal measurable insights into the model's ability to capture and represent the nuances of culinary diversity, while also identifying areas where the model's biases and limitations emerge.

Explaining Explainability: Towards Deeper Actionable Insights into Deep Learning through Second-order Explainability

Explainability plays a crucial role in providing a more comprehensive understanding of deep learning models' behaviour. This allows for thorough validation of the model's performance, ensuring that its decisions are based on relevant visual indicators and not biased toward irrelevant patterns existing in training data. However, existing methods provide only instance-level explainability, which requires manual analysis of each sample. Such manual review is time-consuming and prone to human biases. To address this issue, the concept of second-order explainable AI (SOXAI) was recently proposed to extend explainable AI (XAI) from the instance level to the dataset level. SOXAI automates the analysis of the connections between quantitative explanations and dataset biases by identifying prevalent concepts. In this work, we explore the use of this higher-level interpretation of a deep neural network's behaviour to allows us to "explain the explainability" for actionable insights. Specifically, we demonstrate for the first time, via example classification and segmentation cases, that eliminating irrelevant concepts from the training set based on actionable insights from SOXAI can enhance a model's performance.

ShapeShift: Superquadric-based Object Pose Estimation for Robotic Grasping

Object pose estimation is a critical task in robotics for precise object manipulation. However, current techniques heavily rely on a reference 3D object, limiting their generalizability and making it expensive to expand to new object categories. Direct pose predictions also provide limited information for robotic grasping without referencing the 3D model. Keypoint-based methods offer intrinsic descriptiveness without relying on an exact 3D model, but they may lack consistency and accuracy. To address these challenges, this paper proposes ShapeShift, a superquadric-based framework for object pose estimation that predicts the object's pose relative to a primitive shape which is fitted to the object. The proposed framework offers intrinsic descriptiveness and the ability to generalize to arbitrary geometric shapes beyond the training set.

MMRNet: Improving Reliability for Multimodal Computer Vision for Bin Picking via Multimodal Redundancy

Recently, there has been tremendous interest in industry 4.0 infrastructure to address labor shortages in global supply chains. Deploying artificial intelligence-enabled robotic bin picking systems in real world has become particularly important for reducing labor demands and costs while increasing efficiency. To this end, artificial intelligence-enabled robotic bin picking systems may be used to automate bin picking, but may also cause expensive damage during an abnormal event such as a sensor failure. As such, reliability becomes a critical factor for translating artificial intelligence research to real world applications and products. In this paper, we propose a reliable vision system with MultiModal Redundancy (MMRNet) for tackling object detection and segmentation for robotic bin picking using data from different modalities. This is the first system that introduces the concept of multimodal redundancy to combat sensor failure issues during deployment. In particular, we realize the multimodal redundancy framework with a gate fusion module and dynamic ensemble learning. Finally, we present a new label-free multimodal consistency score that utilizes the output from all modalities to measure the overall system output reliability and uncertainty. Through experiments, we demonstrate that in an event of missing modality, our system provides a much more reliable performance compared to baseline models. We also demonstrate that our MC score is a more powerful reliability indicator for outputs during inference time where model generated confidence score are often over-confident.

MetaGraspNet: A Large-Scale Benchmark Dataset for Scene-Aware Ambidextrous Bin Picking via Physics-based Metaverse Synthesis

Autonomous bin picking poses significant challenges to vision-driven robotic systems given the complexity of the problem, ranging from various sensor modalities, to highly entangled object layouts, to diverse item properties and gripper types. Existing methods often address the problem from one perspective. Diverse items and complex bin scenes require diverse picking strategies together with advanced reasoning. As such, to build robust and effective machine-learning algorithms for solving this complex task requires significant amounts of comprehensive and high quality data. Collecting such data in real world would be too expensive and time prohibitive and therefore intractable from a scalability perspective. To tackle this big, diverse data problem, we take inspiration from the recent rise in the concept of metaverses, and introduce MetaGraspNet, a large-scale photo-realistic bin picking dataset constructed via physics-based metaverse synthesis. The proposed dataset contains 217k RGBD images across 82 different article types, with full annotations for object detection, amodal perception, keypoint detection, manipulation order and ambidextrous grasp labels for a parallel-jaw and vacuum gripper. We also provide a real dataset consisting of over 2.3k fully annotated high-quality RGBD images, divided into 5 levels of difficulties and an unseen object set to evaluate different object and layout properties. Finally, we conduct extensive experiments showing that our proposed vacuum seal model and synthetic dataset achieves state-of-the-art performance and generalizes to real world use-cases.

COVID-Net US-X: Enhanced Deep Neural Network for Detection of COVID-19 Patient Cases from Convex Ultrasound Imaging Through Extended Linear-Convex Ultrasound Augmentation Learning

As the global population continues to face significant negative impact by the on-going COVID-19 pandemic, there has been an increasing usage of point-of-care ultrasound (POCUS) imaging as a low-cost and effective imaging modality of choice in the COVID-19 clinical workflow. A major barrier with widespread adoption of POCUS in the COVID-19 clinical workflow is the scarcity of expert clinicians that can interpret POCUS examinations, leading to considerable interest in deep learning-driven clinical decision support systems to tackle this challenge. A major challenge to building deep neural networks for COVID-19 screening using POCUS is the heterogeneity in the types of probes used to capture ultrasound images (e.g., convex vs. linear probes), which can lead to very different visual appearances. In this study, we explore the impact of leveraging extended linear-convex ultrasound augmentation learning on producing enhanced deep neural networks for COVID-19 assessment, where we conduct data augmentation on convex probe data alongside linear probe data that have been transformed to better resemble convex probe data. Experimental results using an efficient deep columnar anti-aliased convolutional neural network designed via a machined-driven design exploration strategy (which we name COVID-Net US-X) show that the proposed extended linear-convex ultrasound augmentation learning significantly increases performance, with a gain of 5.1% in test accuracy and 13.6% in AUC.

MetaGraspNet: A Large-Scale Benchmark Dataset for Vision-driven Robotic Grasping via Physics-based Metaverse Synthesis

There has been increasing interest in smart factories powered by robotics systems to tackle repetitive, laborious tasks. One impactful yet challenging task in robotics-powered smart factory applications is robotic grasping: using robotic arms to grasp objects autonomously in different settings. Robotic grasping requires a variety of computer vision tasks such as object detection, segmentation, grasp prediction, pick planning, etc. While significant progress has been made in leveraging of machine learning for robotic grasping, particularly with deep learning, a big challenge remains in the need for large-scale, high-quality RGBD datasets that cover a wide diversity of scenarios and permutations. To tackle this big, diverse data problem, we are inspired by the recent rise in the concept of metaverse, which has greatly closed the gap between virtual worlds and the physical world. Metaverses allow us to create digital twins of real-world manufacturing scenarios and to virtually create different scenarios from which large volumes of data can be generated for training models. In this paper, we present MetaGraspNet: a large-scale benchmark dataset for vision-driven robotic grasping via physics-based metaverse synthesis. The proposed dataset contains 100,000 images and 25 different object types and is split into 5 difficulties to evaluate object detection and segmentation model performance in different grasping scenarios. We also propose a new layout-weighted performance metric alongside the dataset for evaluating object detection and segmentation performance in a manner that is more appropriate for robotic grasp applications compared to existing general-purpose performance metrics. Our benchmark dataset is available open-source on Kaggle, with the first phase consisting of detailed object detection, segmentation, layout annotations, and a layout-weighted performance metric script.