Project Website Research Articles

Strategies used for knowledge dissemination and outreach in a renewable energy project

This work describes the actions carried out by participants of the Mexican Center for Innovation in Ocean Energy project (CEMIE-Océano) to disseminate their advances and results, to the general public and to the scientific community. The project was set up to assess the energy potential in Mexican coastal areas related to waves, tides and currents, and salinity and thermal gradients, and also to investigate their expected environmental and social impacts. The main actions by the CEMIE-Océano dissemination, outreach and press area, 2017 - 2023, were the design, construction and administration of the project website, the regular publication of activities and major achievements on social media via Facebook and Twitter, and the editing of 28 scientific books and nine editions of a biannual newsletter. Eight work meetings, three conferences and a range of science dissemination seminars were also organized. With the strategies presented in this work, for a value close to 3% of the total project budget, the participation of all the collaborators was achieved. In addition, the CEMIE-Océano project is widely recognised within the community of scientists from Mexico and Latin America, and others interested in renewable energies and energy transition issues. Keywords: CEMIE-Oceano project, ocean energy, dissemination, outreach.

Laissez-Faire or Sensitive Policymaking: The Legacy of Creative Clusters on Brownfield Sites in Berlin

With his saying “Berlin—poor, but sexy!,” former Governing Mayor Klaus Wowereit framed the motto for a decade of transition in which the German capital became a Mecca for artists, media industries, and creative people from all over the world. Building on a longstanding tradition of tolerating diversity and as a centre of high culture and bohemians, the city developed a new cultural-political identity from a deep transformation crisis after German unification and the extensive loss of its industrial base. In conjunction with a blossoming of temporary uses in a wide variety of vacant properties, often abandoned production, infrastructure, or storage areas, an intense creative scene unfolded. Since the 2010s, this scene has been massively threatened by displacement due to the changed real estate market situation. Over the years, the city has tried to counteract this situation through cultural policy initiatives and niche projects for bottom-up initiatives, with limited success. Against the backdrop of accelerated development of former brownfield sites and funding cuts in urban cultural policy, the question currently arises as to what place subculture can occupy in urban policy in the future. Based on official documents, books, scholarly articles, project websites, newspaper articles, and own observations, this article attempts to evaluate the respective policies in the city over time, to place them in the context of approaches to a more land-security-oriented policy, and to make clear what role the re-used spaces and buildings from the industrial age play in this.

2D Echocardiographic aortic root dimensions and longitudinal analysis with artificial intelligence using an open dataset from a UK-wide collaborative

Abstract Background Surveillance of aortic dimensions requires reproducible measurements, and knowledge of the rate of progression in those with dilatation. Purpose To develop an open machine-learning method for measuring the aortic root and proximal ascending aorta on echocardiograms, validate it through a multi-expert panel from the Unity UK Echocardiography AI Collaborative, and applying the AI to derive the rate of progression in historical databases. Methods The neural network was trained on 1478 parasternal long-axis images. Each image was labelled with key points for the aortic annulus, sinus and sinotubular junction, and proximal ascending aorta dimensions. Labels accommodated inner-to-inner and leading-to-leading conventions. For validation, the end-diastolic and mid-systolic images were identified (accommodating different guidelines recommendations) from 100 studies, and 10 expert echocardiographers made the aortic measurements. The consensus of experts defined the reference standard, and the variation between individual experts defined the acceptable variation for the AI. The AI was then applied to 724 echocardiograms of 102 patients under surveillance spanning 12 years (ranging from 4 to 17 scans per patient) without intervention. Training data labels and networks are made freely available on our project website. Results The median absolute deviations between the AI and the expert consensus ranged from 0.06cm to 0.17cm across the 16 measurements. For the individual expert opinions this was from 0.08cm to 0.19cm. For 7/16 measurements, there was no statistically significant difference between the AI deviation and the individual-expert deviation. The AI’s confidence level was a useful indicator of the reliability of the AI measure. For the top 9 deciles of AI confidence level, the AI measurements had significantly smaller deviations than the individual experts. The rates of progression averaged across the three aortic measurements was 2.0 mm/decade (95% CI 1.0 to 3.0; p =0.007). Conclusion Expert consensus can be used to both define the reference standard and also the range of acceptable deviation from consensus. Overall, the AI performs similarly to experts and more importantly, provides an automated estimate of confidence. The validated AI is consistent over time, and when applied retrospectively to patients with dilated aortic roots, found the rate of progression was more rapid than in previously reported healthy patients. This approach could be applied generally in developing medical AI to not only make reproducible measurements but to automate large scale longitudinal studies with little input, and ultimately serves as a useful research and clinical tool.Figure: AI(red) vs Expert (blue)Table: AIDeviations from Consensus

The SocialAI school: a framework leveraging developmental psychology toward artificial socio-cultural agents.

Developmental psychologists have long-established socio-cognitive abilities as fundamental to human intelligence and development. These abilities enable individuals to enter, learn from, and contribute to a surrounding culture. This drives the process of cumulative cultural evolution, which is responsible for humanity's most remarkable achievements. AI research on social interactive agents mostly concerns the emergence of culture in a multi-agent setting (often without a strong grounding in developmental psychology). We argue that AI research should be informed by psychology and study socio-cognitive abilities enabling to enter a culture as well. We draw inspiration from the work of Michael Tomasello and Jerome Bruner, who studied socio-cognitive development and emphasized the influence of a cultural environment on intelligence. We outline a broader set of concepts than those currently studied in AI to provide a foundation for research in artificial social intelligence. Those concepts include social cognition (joint attention, perspective taking), communication, social learning, formats, and scaffolding. To facilitate research in this domain, we present The SocialAI school-a tool that offers a customizable parameterized suite of procedurally generated environments. This tool simplifies experimentation with the introduced concepts. Additionally, these environments can be used both with multimodal RL agents, or with pure-text Large Language Models (LLMs) as interactive agents. Through a series of case studies, we demonstrate the versatility of the SocialAI school for studying both RL and LLM-based agents. Our motivation is to engage the AI community around social intelligence informed by developmental psychology, and to provide a user-friendly resource and tool for initial investigations in this direction. Refer to the project website for code and additional resources: https://sites.google.com/view/socialai-school.

Empowering Agrifood System with Artificial Intelligence: A Survey of the Progress, Challenges and Opportunities

With the world population rapidly increasing, transforming our agrifood systems to be more productive, efficient, safe, and sustainable is crucial to mitigate potential food shortages. Recently, artificial intelligence (AI) techniques such as deep learning (DL) have demonstrated their strong abilities in various areas, including language, vision, remote sensing (RS), and agrifood systems applications. However, the overall impact of AI on agrifood systems remains unclear. In this paper, we thoroughly review how AI techniques can transform agrifood systems and contribute to the modern agrifood industry. Firstly, we summarize the data acquisition methods in agrifood systems, including acquisition, storage, and processing techniques. Secondly, we present a progress review of AI methods in agrifood systems, specifically in agriculture, animal husbandry, and fishery, covering topics such as agrifood classification, growth monitoring, yield prediction, and quality assessment. Furthermore, we highlight potential challenges and promising research opportunities for transforming modern agrifood systems with AI. We hope this survey could offer an overall picture to newcomers in the field and serve as a starting point for their further research. The project website is https://github.com/Frenkie14/Agrifood-Survey.

Active Learning for Deep Visual Tracking.

Convolutional neural networks (CNNs) have been successfully applied to the single target tracking task in recent years. Generally, training a deep CNN model requires numerous labeled training samples, and the number and quality of these samples directly affect the representational capability of the trained model. However, this approach is restrictive in practice, because manually labeling such a large number of training samples is time-consuming and prohibitively expensive. In this article, we propose an active learning method for deep visual tracking, which selects and annotates the unlabeled samples to train the deep CNN model. Under the guidance of active learning, the tracker based on the trained deep CNN model can achieve competitive tracking performance while reducing the labeling cost. More specifically, to ensure the diversity of selected samples, we propose an active learning method based on multiframe collaboration to select those training samples that should be and need to be annotated. Meanwhile, considering the representativeness of these selected samples, we adopt a nearest-neighbor discrimination method based on the average nearest-neighbor distance to screen isolated samples and low-quality samples. Therefore, the training samples' subset selected based on our method requires only a given budget to maintain the diversity and representativeness of the entire sample set. Furthermore, we adopt a Tversky loss to improve the bounding box estimation of our tracker, which can ensure that the tracker achieves more accurate target states. Extensive experimental results confirm that our active-learning-based tracker (ALT) achieves competitive tracking accuracy and speed compared with state-of-the-art trackers on the seven most challenging evaluation benchmarks. Project website: https://sites.google.com/view/altrack/.

FMB: A functional manipulation benchmark for generalizable robotic learning

In this paper, we propose a real-world benchmark for studying robotic learning in the context of functional manipulation: a robot needs to accomplish complex long-horizon behaviors by composing individual manipulation skills in functionally relevant ways. The core design principles of our Functional Manipulation Benchmark (FMB) emphasize a harmonious balance between complexity and accessibility. Tasks are deliberately scoped to be narrow, ensuring that models and datasets of manageable scale can be utilized effectively to track progress. Simultaneously, they are diverse enough to pose a significant generalization challenge. Furthermore, the benchmark is designed to be easily replicable, encompassing all essential hardware and software components. To achieve this goal, FMB consists of a variety of 3D-printed objects designed for easy and accurate replication by other researchers. The objects are procedurally generated, providing a principled framework to study generalization in a controlled fashion. We focus on fundamental manipulation skills, including grasping, repositioning, and a range of assembly behaviors. The FMB can be used to evaluate methods for acquiring individual skills, as well as methods for effectively combining and ordering such skills in order to solve complex, multi-stage manipulation tasks. We also offer an imitation learning framework that includes a suite of policies trained to solve the proposed tasks. This enables researchers to utilize our tasks as a versatile toolkit for examining various parts of the pipeline. For example, researchers could propose a better design for a grasping controller and evaluate it in combination with our baseline reorientation and assembly policies as part of a pipeline for solving multi-stage tasks. Our dataset, object CAD files, code, and evaluation videos can be found on our project website: https://functional-manipulation-benchmark.github.io .

Experiences of Acquired Brain Injury as Expressed Through Mask-Making: A Qualitative Analysis.

Acquired brain injury (ABI) extends beyond the immediate aftermath, and understanding individual experience of ABI is paramount to providing effective support mechanisms. This study examined how people with ABI used mask-making to engage in self-expression. Publicly available data from the Unmasking Brain Injury Project website, an advocacy group for people with ABI, were analyzed. A qualitative approach with hybrid inductive/deductive analysis was used to explore how people with ABI use art to express their emotional experiences of ABI. In total, 1,049 masks had narratives describing the visual components and meaning in the associated masks. Three major themes emerged: the multiplicity of experiences after ABI, including positive, negative, and mixed outcomes; the expression of emotional pain and living with loss through art; and the importance of positivity, purpose, and faith in the experience with ABI. Multiplicity was the central expression represented in the masks. The most commonly represented emotional experiences were: emotional pain and living with loss and positivity, purpose, and faith. The masks incorporated literal depictions of the ABI, visual metaphors, and intentional use of colors and shapes for self-expression. These results support previous findings that emotional experiences after ABI are complex and often interlaced with several emotions. The most prominent feelings include emotional pain and living with loss, and positivity, purpose, and faith. Mask-making may be a particularly appropriate method for people with ABI to process or express emotional experiences, as well as to provide clinicians with visual records of sense of self, coping, or progress.

Using a negative spatial auto-correlation index to evaluate and improve intrinsic tag map’s multi-scale visualization capabilities

ABSTRACT The popularity of tag clouds has sparked significant interest in the geographic research community, leading to the development of map-based adaptations known as intrinsic tag maps. However, existing methodologies for tag maps primarily focus on tag layout at specific scales, which may result in large empty areas or close proximity between tags when navigating across multiple scales. This issue arises because initial tag layouts may not ensure an even distribution of tags with varying sizes across the region. To address this problem, we incorporate the negative spatial auto-correlation index into tag maps to assess the uniformity of tag size distribution. Subsequently, we integrate this index into a TIN-based intrinsic tag map layout approach to enhance its ability to support multi-scale visualization. This enhancement involves iteratively filtering out candidate tags and selecting optimal tags that meet the defined index criteria. Experimental findings from two representative areas (the USA and Italy) demonstrate the efficacy of our approach in enhancing multi-scale visualization capabilities, albeit with trade-offs in compactness and time efficiency. Specifically, when retaining the same number of tags in the layout, our approach achieves higher compactness but requires more time. Conversely, when reducing the number of tags in the layout, our approach exhibits reduced time requirements but lower compactness. Furthermore, we discuss the effectiveness of various applied strategies aligned with existing approaches to generate diverse intrinsic tag maps tailored to user preferences. Additional details and resources can be found on our project website: https://github.com/TrentonWei/Multi-scale-TagMap.

Delving into CLIP latent space for Video Anomaly Recognition

We tackle the complex problem of detecting and recognising anomalies in surveillance videos at the frame level, utilising only video-level supervision. We introduce the novel method AnomalyCLIP, the first to combine Vision and Language Models (VLMs), such as CLIP, with multiple instance learning for joint video anomaly detection and classification. Our approach specifically involves manipulating the latent CLIP feature space to identify the normal event subspace, which in turn allows us to effectively learn text-driven directions for abnormal events. When anomalous frames are projected onto these directions, they exhibit a large feature magnitude if they belong to a particular class. We also leverage a computationally efficient Transformer architecture to model short- and long-term temporal dependencies between frames, ultimately producing the final anomaly score and class prediction probabilities. We compare AnomalyCLIP against state-of-the-art methods considering three major anomaly detection benchmarks, i.e. ShanghaiTech, UCF-Crime, and XD-Violence, and empirically show that it outperforms baselines in recognising video anomalies. Project website and code are available at https://lucazanella.github.io/AnomalyCLIP/.

Alternative education in Russian megacities: a chance for a different school

Introduction. The relevance of the study is determined by the search for innovative ways to develop school education. The study of the history and current state of alternative education in the largest Russian cities plays a significant role in this search. Systematisation of private educational projects will help families to better orientate themselves in alternative education and avoid its destructive models. Materials and Methods. An analysis of scientific literature was conducted, and information from the websites of 665 private educational projects was collected and analyzed, including 322 from Moscow, 183 from Saint Petersburg, 57 from Yekaterinburg, and 51 each from Novosibirsk and Kazan. In-depth interviews with their leaders were conducted, focus groups with experts were held, and an analysis of federal state statistics was made. Research Results. The coverage of children by humanistic alternative education practices in Russia is less than 1%. However, over 10 years, there has been a noticeable growth in the number of educational projects outside the system. An analysis of Project websites showed that only 8% of them consider themselves to be alternative education. 43% of Projects have a license for primary educational activities. Most often, projects implement primary general education programs. 23% of projects claim to implement proprietary programs. A small number of projects are based on specific didactic systems, approaches and practices. The typology of Alternative Education Projects is made taking into account the role of the child in the formation and implementation of educational requests. The study identified three spaces of alternative education: projects supporting family education, projects with established didactic systems (program schools), and A-schools. There are destructive practices in alternative education. Conclusion. Alternative education in the large cities of Russia is diverse and includes both humanistic and other pedagogical practices, as well as a degree of uncertainty. The pedagogical practices of alternative education are difficult to typologize. This living organism continuously transforms, sensitively responding to the external environment.

Stick to your role! Stability of personal values expressed in large language models.

The standard way to study Large Language Models (LLMs) through benchmarks or psychology questionnaires is to provide many different queries from similar minimal contexts (e.g. multiple choice questions). However, due to LLM's highly context-dependent nature, conclusions from such minimal-context evaluations may be little informative about the model's behavior in deployment (where it will be exposed to many new contexts). We argue that context-dependence should be studied as another dimension of LLM comparison alongside others such as cognitive abilities, knowledge, or model size. In this paper, we present a case-study about the stability of value expression over different contexts (simulated conversations on different topics), and as measured using a standard psychology questionnaire (PVQ) and behavioral downstream tasks. We consider 21 LLMs from six families. Reusing methods from psychology, we study Rank-order stability on the population (interpersonal) level, and Ipsative stability on the individual (intrapersonal) level. We explore two settings: with and without instructing LLMs to simulate particular personalities. We observe similar trends in the stability of models and model families-Mixtral, Mistral, GPT-3.5 and Qwen families being more stable than LLaMa-2 and Phi-over those two settings, two different simulated populations, and even on three downstream behavioral tasks. When instructed to simulate particular personas, LLMs exhibit low Rank-Order stability, and this stability further diminishes with conversation length. This highlights the need for future research directions on LLMs that can coherently simulate a diversity of personas, as well as how context-dependence can be studied in more thorough and efficient ways. This paper provides a foundational step in that direction, and, to our knowledge, it is the first study of value stability in LLMs. The project website with code is available at https://sites.google.com/view/llmvaluestability.

A study exploring the causal relationship between glaucoma and anxiety disorders.

Glaucoma, a leading cause of global blindness, is characterized by optic nerve damage and visual field loss. Previous studies have suggested a potential association between glaucoma and anxiety disorders. However, the causal relationship between these two conditions remains unclear. In this study, we conducted a Mendelian Randomization analysis to investigate the causal relationship between glaucoma and anxiety disorders. We sourced Genome-Wide Association Study (GWAS) datasets for glaucoma and anxiety with the largest sample sizes from the Integrative Epidemiology Unit OpenGWAS (IEU OpenGWAS) project website. Instrumental variables were selected based on specific criteria, and statistical analyses were performed using the R programming language. After filtering and merging the datasets, a total of 60 Single Nucleotide Polymorphisms (SNPs) were obtained for analysis. Regression models were applied to assess the causal relationship between glaucoma and anxiety disorders. The results from all four methods indicated that glaucoma does not cause anxiety disorders (p > 0.05). Through rigorous Mendelian Randomization analysis, our findings indicate that glaucoma is not a causative factor for anxiety, with minimal influence from confounding factors in this study. These findings enhance our understanding of the relationship between glaucoma and anxiety.

“Equality, freedom, solidarity – the issues on which I was raised”

As part of this special issue, we received this reflective piece from Viena Chilena 73 | 23, a project initiated to record and articulate the history and memory of Chilean exile in Austria and Austrian-Chilean solidarity during the 50 years since the 1973 military coup in Chile. This contribution pieces together elements of the project website, extracts of their archival work and reflections from the project collective.

SMERF: Streamable Memory Efficient Radiance Fields for Real-Time Large-Scene Exploration

Recent techniques for real-time view synthesis have rapidly advanced in fidelity and speed, and modern methods are capable of rendering near-photorealistic scenes at interactive frame rates. At the same time, a tension has arisen between explicit scene representations amenable to rasterization and neural fields built on ray marching, with state-of-the-art instances of the latter surpassing the former in quality while being prohibitively expensive for real-time applications. We introduce SMERF, a view synthesis approach that achieves state-of-the-art accuracy among real-time methods on large scenes with footprints up to 300 m 2 at a volumetric resolution of 3.5 mm 3 . Our method is built upon two primary contributions: a hierarchical model partitioning scheme, which increases model capacity while constraining compute and memory consumption, and a distillation training strategy that simultaneously yields high fidelity and internal consistency. Our method enables full six degrees of freedom navigation in a web browser and renders in real-time on commodity smartphones and laptops. Extensive experiments show that our method exceeds the state-of-the-art in real-time novel view synthesis by 0.78 dB on standard benchmarks and 1.78 dB on large scenes, renders frames three orders of magnitude faster than state-of-the-art radiance field models, and achieves real-time performance across a wide variety of commodity devices, including smartphones. We encourage readers to explore these models interactively at our project website: https://smerf-3d.github.io.

Dotting the i’s and crossing the t’s: Operationalising a teaching and learning community of practice

Social theories of learning propose that communities serve as vital building blocks for knowledge sharing and construction (Bandura, 1977; Cobb, 1994). In essence, learning thrives when individuals congregate in a community where they exchange tacit knowledge and expose areas of ignorance (Wenger et al., 2002). Nevertheless, the process of cultivating a community of practice is not necessarily straightforward, as it requires meticulous planning and decision-making. This instrumental case study seeks to explicate the formation and development of a teaching and learning community of practice (T&L CoP) at a research-intensive university in Hong Kong. More specifically, it explores how the structural model of the community of practice – domain, community, and practice – are applied and manifested. Multiple sources of documents, including a project proposal, a project website, a project progress report and archival records, were analysed using deductive thematic analysis. The results reveal that the T&L CoP is constitutive of an overarching domain of teaching and learning, within which ten special interest groups are formed. The T&L CoP also bring together 150 professoriate and professional staff from across eight faculties. The community facilitates knowledge exchange through periodic professional development workshops and materials as well as sharing sessions that highlight exemplary teaching practice. The study concludes by discussing three fundamental questions regarding the operationalisation of a CoP, in terms of expansion, retention, as well as sustainability, and argues that the T&L CoP helps cultivate members’ social capital through improved networks and visibility among the teaching and learning community.

Bone health education in individuals with spinal cord injury or disease-the Bare Bones Podcast Series: plan it, produce it, post it!

The Consortium for Spinal Cord Medicine's inaugural Clinical Practice Guideline for Bone Health and Osteoporosis Management for Individuals with Spinal Cord Injury or Disease (CSCM-CPG) was published in 2022 for a clinician audience. The aim of this project was to develop a podcast series to ensure people with lived experience with Spinal Cord Injury or Disease (PLEX) understand the CSCM-CPG content and know how to act to reduce their fracture risk. The "Bare Bones Podcast Series" consists of nine episodes; one related to each CSCM-CPG chapter. The podcast content and the questions asked in each podcast were co-developed by PLEX partners (PLEX-P) and the project team. Two PLEX-P acted as co-hosts for the series. The invited speaker(s) were CSCM-CPG expert panel members who participated in an informal dialogue with the hosts. Each podcast closes with a specific action a listener can do to advance their bone health. The related Educational Action Planning Tool (EAT) handouts contain text and infographic information specific to each podcast episode and include key concepts and a specific actionable take-home message. Local PLEX reviewers (PLEX-R) were invited to review podcast episodes and EATs and provide their feedback through focus group participation or one-on-one (1:1) interviews. The project team revised the podcast episodes and the EATs based on feedback from the PLEX-R prior to releasing them online. Nine podcast episodes and related EATs were designed and created collaboratively with 3 PLEX-P, 22 PLEX-R, 11 CSCM-CPG expert panel members, and the project team. The episodes were titled: "Introduction to the Bare Bones of Bone Health"; "Fracture 101"; "Blood Tests-a Window into You"; "I See Your Skeleton"; "Vitamin D for all, Calcium for Some"; "Get Moving and Loading"; "Pills or Poisons & Atomic Habits"; "Snap and Crack"; and "Directions for Research". The Bare Bones Podcast Series was shared through the project website. The podcasts will aid PLEX and their family caregivers to advocate for ongoing bone health assessments and to promote an ongoing dialogue with care team members regarding how to prevent fractures and fracture-related morbidity and mortality.

An evaluation of CNN models and data augmentation techniques in hierarchical localization of mobile robots

This work presents an evaluation of CNN models and data augmentation to carry out the hierarchical localization of a mobile robot by using omnidirectional images. In this sense, an ablation study of different state-of-the-art CNN models used as backbone is presented and a variety of data augmentation visual effects are proposed for addressing the visual localization of the robot. The proposed method is based on the adaption and re-training of a CNN with a dual purpose: (1) to perform a rough localization step in which the model is used to predict the room from which an image was captured, and (2) to address the fine localization step, which consists in retrieving the most similar image of the visual map among those contained in the previously predicted room by means of a pairwise comparison between descriptors obtained from an intermediate layer of the CNN. In this sense, we evaluate the impact of different state-of-the-art CNN models such as ConvNeXt for addressing the proposed localization. Finally, a variety of data augmentation visual effects are separately employed for training the model and their impact is assessed. The performance of the resulting CNNs is evaluated under real operation conditions, including changes in the lighting conditions. Our code is publicly available on the project website https://github.com/juanjo-cabrera/IndoorLocalizationSingleCNN.git.

Reporting quality of published reviews of commercial and publicly available mobile health apps (mHealth app reviews): a scoping review protocol

IntroductionReviews of commercial and publicly available smartphone (mobile) health applications (mHealth app reviews) are being undertaken and published. However, there is variation in the conduct and reporting of mHealth app...

Performing Hunger in Taiwan: Images, Objects, and Reflections

In 1982, Hungarian-American artist Agnes Denes planted and harvested two acres of wheat on the Battery Park landfill, Manhattan, to draw attention to mismanagement, waste, and world hunger. More recently, from 2014 to the present, American environmental activist Rob Greenfield addressed food waste and demonstrated food freedom by dumpster diving and foraging. The ethos of these two examples align with Environmental Art, a general education course that I teach at Chung Yuan Christian University (CYCU), Taiwan. Environmental Art is an umbrella term for art that interacts with or comments on the environment. Among its subgenres, Ecological Art emphasizes the preservation and remediation of life, resources, and ecosystems of Earth. This article explores the online teaching pedagogy for the Environmental Art course final assignment, The Hunger Project, and its ramifications in relation to performance. Using Ecological Art and hunger statistics as a springboard, students examine hunger in Taiwan and in the global context. They consider various dimensions of hunger, such as access and distribution of food, and the question: why do the lack and the excess of food coexist? Due to the COVID-19 pandemic, I and students realized the entire “The Hunger Project” online. The pedagogical methods facilitated the “The Hunger Project Website,” where students staged texts, images, and objects to visualize their reflections on hunger. Working in groups, students addressed metaphorical “hungers”, such as the hunger for being healthy, linked social class to food consumption, and proposed education as an antidote to poverty and food deficiency. This essay considers the performative and educational significance of this learning process in relation to hunger action.