This work demonstrates the implementation and use of an encoder-decoder model to perform a many-to-many mapping of video data to text captions. The many-to-many mapping occurs via an input temporal sequence of video frames to an output sequence of words to form a caption sentence. Data preprocessing, model construction, and model training are discussed. Caption correctness is evaluated using 2-gram BLEU scores across the different splits of the dataset. Specific examples of output captions were shown to demonstrate model generality over (...) the video temporal dimension. Predicted captions were shown to generalize over video action, even in instances where the video scene changed dramatically. Model architecture changes are discussed to improve sentence grammar and correctness. (shrink)

Image captioning has seen remarkable advancements with the integration of deep learning techniques, notably Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, for generating descriptive captions for images. Despite these improvements, capturing intricate details and context remains a challenge. This project introduces an enhanced image captioning model that integrates transformers with an attention mechanism to address these limitations. By leveraging CNNs for feature extraction and LSTMs for sequence generation, while utilizing transformers to apply sophisticated attention to (...) significant image regions, the proposed model aims to generate more contextually rich and coherent captions. Experimental results indicate that incorporating transformers with attention mechanisms leads to a significant enhancement in caption accuracy and descriptiveness, surpassing traditional CNN-LSTM models. This advancement is particularly beneficial in various applications, including assistive technologies for the visually impaired, content-based image retrieval systems, automatic image annotation for digital asset management, and improved human-computer interaction. This approach represents a substantial step forward in achieving more precise and detailed image captioning, with potential impacts across numerous fields. (shrink)

This work demonstrates the implementation and use of an encoder-decoder model to perform a many-to-many mapping of video data to text captions. The many-to-many mapping occurs via an input temporal sequence of video frames to an output sequence of words to form a caption sentence. Data preprocessing, model construction, and model training are discussed. Caption correctness is evaluated using 2-gram BLEU scores across the different splits of the dataset. Specific examples of output captions were shown to demonstrate model generality over (...) the video temporal dimension. Predicted captions were shown to generalize over video action, even in instances where the video scene changed dramatically. Model architecture changes are discussed to improve sentence grammar and correctness. (shrink)

This work demonstrates the implementation and use of an encoder-decoder model to perform a many-to-many mapping of video data to text captions. The many-to-many mapping occurs via an input temporal sequence of video frames to an output sequence of words to form a caption sentence. Data preprocessing, model construction, and model training are discussed. Caption correctness is evaluated using 2-gram BLEU scores across the different splits of the dataset. Specific examples of output captions were shown to demonstrate model generality over (...) the video temporal dimension. Predicted captions were shown to generalize over video action, even in instances where the video scene changed dramatically. Model architecture changes are discussed to improve sentence grammar and correctness. (shrink)

The concept of 'authenticity' is highly valued on social media sites (SMSes), despite its ambiguous nature and definition. One interpretation of 'authenticity' by media scholars is a human's congruence with online portrayals of themselves (e.g. posting spontaneous photographs from their lives, or using real biodata online). For marketers and 'influencers', these patterns of behaviour can achieve certain gains: sales for a business, or success of a campaign. For existentialist philosophers, using 'authenticity' as a means to an end is against its (...) very definition. In this paper, I investigate what SMS users are looking for by their supposed 'authentic' portrayal online. My experimental approach draws upon empirical data from the Instagram social media site. Using machine learning techniques, descriptions and features of posts - including subjects, captions, and contexts - will be categorised and aggregated. I will then interpret these findings, drawing upon work by Taylor, Golomb, and Guignon, whose works on authenticity are based on mid-20th century existentalists. I argue that the existentialist ideals on authenticity are not necessarily present in contemporary SMS use. I will also argue that the popular interpretation of authenticity on SMSes can be self-defeating, when it seeks to turn the 'for-itself' into an 'in-itself'. (shrink)

A videographic study of origami is presented in which subjects were observed making four different origami objects under five modes of instruction: photos + captions, illustrations-only, illustrations with small captions, illustrations with large captions, and text-only as control. The objective of the study was to explore the gestures and other actions that subjects produce as they try to follow instructions rather than to determine the most effective style of instruction per se. We found that the task of situating instructions to (...) the context at hand is error prone and that to facilitate it subjects gesture, point, re-orient illustrations, and generally do things that have no function other than to change the epistemic and interactive landscape of activity so they can more easily understand what is to be done. These studies bear on the new questions designers are asking about the placement, timing, and pace of instructions that digital aids now provide and on the fundamental question of how humans embed themselves in an activity by framing their task in a situation specific manner. (shrink)

"Philosophy of Happiness: A Critical Introduction" summarizes (a) what philosophy of happiness is, (b) why it should matter to us, (c) what assistance we can draw from philosophy, empiric science, religion, and self-help sources, and (d) why taking an independent approach is both necessary and feasible. -/- The article is in PDF format, 60 pages. The table of contents links directly to the listed captions. Also available in an html version under the phone variant of the referenced philosophy of happiness (...) website. -/- Table of Contents -/- 1. The Definition of Happiness. -/- 2. The Definition of Philosophy of Happiness. -/- 3. Taking Refuge from the Complexity of Happiness. -/- 4. Self-Help Contributions to Philosophy of Happiness. -/- 5. Religious Contributions to Philosophy of Happiness. -/- 6. Philosophical Authorities on Philosophy of Happiness. -/- 7. Happiness Studies by Sciences Beyond Philosophy. -/- 8. Independent Approaches Toward a Philosophy of Happiness. -/- 9. Further Reading. (shrink)

This Article explores the interpretation and construction of executive orders using as examples President Trump’s two executive orders captioned “Protecting the Nation From Foreign Terrorist Entry Into the United States” (the “Two Executive Orders”). President Trump issued the Two Executive Orders in the context of (among other things) Candidate Trump’s statements such as: “Islam hates us,” and “[W]e can’t allow people coming into this country who have this hatred.” President Trump subsequently provided further context including his tweet about the second (...) of his Two Executive Orders: “People, the lawyers and the courts can call [the second of the Two Executive Orders] whatever they want, but I am calling it what we need and what it is, a TRAVEL BAN!” Although President Trump replaced the first of the Two Executive Orders with the second one and although the Supreme Court by orders dated October 10, 2017 and October 24, 2017, vacated and remanded litigation involving the second order on grounds of mootness, the Two Executive Orders remain highly instructive for those who would understand the interpretation and construction of executive orders. This article therefore examines in detail the original speaker's (i.e., President Trump's) intended meaning and effect of the Two Executive Orders. It performs such examination using insights from the semiotic subfield of pragmatics, a semiotic subfield which explores how real-world people actually use, interpret, and construe language in various real-world contexts (including contexts where the issuer of the Two Executive Orders himself has claimed that “Islam hates us” and has tweeted “TRAVEL BAN!”). Using such insights of pragmatics, this Article also explores why reasonable judges thoroughly versed in legal theory, legal practice, and pragmatics should conclude that President Trump unlawfully targeted Muslims in the Two Executive Orders. This Article, among other things, also questions the sensibility of such notions as “facial legitimacy” to the extent such notions suggest text has meaning apart from context. Keywords: executive orders, speaker meaning, interpretation, construction, semiotics, pragmatics, originalism, speech acts, context, facial legitimacy, Constitutionality, First Amendment,Trump v. Int’l Refugee Assistance Project, Scalia, immigration, semantics, original meaning, travel ban, Trump, textualism. (shrink)

Publication Zero Point shows the results of a four-years inquiry into the concept of 'measurability'. The book encompasses thirty-eight pictureworks and a philosophical thought experiment. It invites the reader to contemplate and compile their personal 'zero point' as a portrait of God. -/- Measurability appears to be a problematic concept. The boundaries between fixed and fluid; between sharp and vague; between coloured and non-coloured; between love and hate. When does one state of affairs turn into the other? When is something (...) still measurable? And: is it essential for something to be measurable at all? -/- A single drinking glass was the starting point for transformational processes of photographic images into multi-layered pictureworks. Pictorial spaces that are variable and subject to chance in an experimental, unconscious method of choice, chance, inspiration, and demolition. The results of split, remodel, and repeat invite you to explore and contemplate the notions of space and time as a dimension, in which meaning remains something definitely unfinished, but in which measurability is crucial. After all, when there is nothing to hold on to, you have to choose a point of departure from which you can reorganise your life. -/- There are several ways to leaf through the book and engage in its content. Each of the thirty-eight pictureworks is accompanied by a philosophical caption. These captions are part of a larger philosophical deliberation of three countenances which together make a portrait of God. Each caption can be read separately next to the picturework. Each picturework can be enjoyed next to a specific caption. Another way to work with the captions is to read them within their larger con text: unfold the last page of the book, reflect, and search for cross-references next to any particular picturework. (shrink)