AI synthetic Touhou fan-art: Aya Shameimaru, Megumu Iizunamaru, Hatate Himekaidou water-playing at the river of Yokai mountain - Tool: ChatGPT plus

The heatwave effect still prolongs here, and I have been too tired of my main research task to anything productive other than very essential now. 

Yet, I still want to keep posting my Touhou Project fan-art as the Touhou Project popularity poll is going to be open from 21 to 28 August this year.

Then, I have asked ChatGPT plus to redraw one of my old illustration following the prompt shown below. 

Aya Shameimaru, Megumu Iizunamaru & Hatate Himekaidou

 

Click the image to enlarge.

The image is posted on my pixiv page:https://www.pixiv.net/en/artworks/148487348

Original image:
Pixiv — Artwork 121317998
Aya Shameimaru, Megumu Iizunamaru & Hatate — Original Blogger Post

 

---

Mountain Stream Tengu Illustration Prompt

Overall Art Direction

Redraw and refine the supplied illustration as a highly polished contemporary Japanese anime illustration with luminous, richly rendered colouring, delicate linework, soft dimensional shading, and an exceptionally refined digital-painting finish.

Use the attached image as the main visual basis and preserve its:

• overall mountain-stream setting

• three-character arrangement

• left-to-right storytelling

• pose relationships

• facial expressions

• colour atmosphere

• water rendering

• lighting direction

• dynamic energy

• humorous mood

• camera distance

• approximate character scale

• overall spatial staging

Keep the image as close as possible to the attached version while correcting the remaining anatomical and design issues described below.

Render all three figures clearly as adult winged mountain-spirit women with mature proportions and a clean, polished contemporary anime finish.

Their designs should remain originalised folklore-inspired fantasy designs rather than explicitly named copyrighted characters.

Maintain the original visual concept and humorous character interaction while improving:

• anatomy

• proportions

• facial construction

• hands

• limbs

• hair flow

• wings

• swimwear construction

• accessories

• lighting

• skin rendering

• water effects

• environmental detail

wherever necessary for a more natural, balanced, and professionally illustrated result.

Allow small adjustments to posture, limb placement, hair, wings, clothing, and accessories where necessary to correct anatomy or strengthen visual balance, but do not fundamentally alter:

• the original scene

• left-to-right character arrangement

• storytelling

• camera placement

• pose relationships

• existing figure-size relationships

Do not introduce exaggerated perspective, dramatic wide-angle distortion, or major character repositioning simply to create additional depth.

Preserve the source image's moderate perspective and character-scale balance:

left reporter slightly closer → centre commander at the primary middle distance → right photographer modestly farther back

The differences in apparent scale should remain restrained and consistent with the reference.

Use smooth and sophisticated contemporary anime rendering with:

• carefully controlled line weight

• thin-to-medium tapered outlines

• subtle painterly gradients

• clean colour transitions

• soft ambient occlusion

• delicate reflected colour

• finely layered highlights

• softly dimensional form modelling

• controlled environmental reflections

• luminous warm skin tones

• transparent water droplets

• damp hair highlights

• warm sunlight interacting with cool stream reflections

Avoid excessively flat cel shading.

Avoid heavy photorealism.

The characters should possess a gentle three-dimensional presence while remaining unmistakably anime in design.

A major visual priority should be the interaction between:

soft luminous skin + transparent water droplets + bright summer sunlight + damp hair + reflected blue-green stream light

RIP Bonnie Tyler - Holding out of a Hero

 

RIP Bonnie Tyler. I love your song "Holding out of a Hero (♬ I need a hero! ♬" which is considered as one of my theme songs! The history moves on as the legend in the golden age of the Western pop music has passed away.  

AI synthetic Touhou fan-art: Yuyuko Saigyoji in Hakugyokurou - Tool: ChatGPT plus

I am still tremendously busy with my current main researcher job as I am in the middle of the research project under a strict supervision for the time being. 

Also, the heatwave has come again, which is currently draining my energy and motivation for many thing on the top of it. 

Therefore, I am playing with the AI again to make the synthetic art experiment.  

As the Touhou Project popularity poll is going to be open from 21 to 28 August this year, I want to advertise for one of my favourite characters I will vote, Yuyuko Saigyouji together with my top favourite Myon-chan (Youmu Konpaku).  I want to support both characters from Hakugyokurou to increase and sustain their popularity! 

Yuyuko Saigyouji ("Yuyu-sama")

Click the image to enlarge.
The image is posted on my pixiv page: https://www.pixiv.net/en/artworks/148439873

Original image:
Pixiv — Artwork 124575093
Yuyuko-sama — Posing for Modelling

---

Elegant Pink-Haired Ghostly Noblewoman Illustration Prompt

Overall Art Direction

Redraw this illustration as a polished early-to-mid-1990s Japanese fantasy RPG key illustration, reflecting the elegant visual language of Japanese fantasy strategy games, anime illustration books, and romantic shōjo-fantasy artwork from approximately 1991–1995.

Improve the line quality, anatomy, proportions, and shading while preserving the original composition.

Keep the original background, colour scheme, character pose, figure placement, and overall composition intact.

The finished artwork should feel like a refined Japanese fantasy game illustration or premium anime visual collection piece from the early-to-mid 1990s, with elegant linework, softly dimensional colouring, and a slightly romantic shōjo-fantasy atmosphere.

---

Facial Design

The facial rendering should lean slightly toward a graceful shōjo-fantasy sensibility.

Do not render the eyes in an excessively sharp, severe, or strongly shōnen-manga style.

Keep the eyes elegant, luminous, softly shaped, and emotionally readable.

Use delicate facial features, graceful eyelashes, softly layered irises, and controlled highlights without making the eyes oversized or excessively modern in appearance.

Her expression should appear graceful, serene, mysterious, and gently playful.

---

Canvas and Aspect Ratio

Use a landscape composition with a width-to-height aspect ratio of 1.414:1, matching the international A-series paper proportion and suitable for high-quality A3 or A4 printing.

My pencil drawing of Myonchan (Youmu Konpaku) - Touhou Project fan-art.

 

Posted on my pixiv page: https://www.pixiv.net/en/artworks/148229878
My pencil drawing of Myonchan (Youmu Konpaku) - Touhou Project fan-art.
I'm voting for her in my top pick of Touhou Project popularity poll coming this month, this year too.

After this experiment, I have recently started the tablet compatible with CLIP STUDIO PAINT again: There are some manoeuvres capable with my direct pencil input. But, the problem is the cost and the possibility to get the one I want. Also, my drawing quality varies depending on my condition too.

Well, I enjoy using the pen-tablet, but I sometimes face its limitations of drawing lines, e.g. hatching, and brush manoeuvre. After having your feedback, I may consider investing for the graphic tablet with screen at some point.

Yet, I am less likely to afford to purchase and spare enough time for playing with the screen tablet this year. My current strategy is continuing with analogue drawing for scenery sketching and the digital painting with the current environment with the pen tablet for something more easily manoeuvre drawings.

 

 

 

 

AI synthetic Touhou fan-art: Youmu and Sanae standing against Tojiko, Futo, and Futo in Touhou 13 ~ Ten Desires - Tool: ChatGPT plus

 

Posted on my pixiv page: https://www.pixiv.net/en/artworks/147796476

YouTube: https://www.youtube.com/shorts/4Htb42KpQzk
NicoNico: https://www.nicovideo.jp/shorts/ss46607508

Lately, I’ve been busy with my main job and have no air conditioning in my room, making it hard to focus on creative work due to my primary research duties. Today, I scanned a rough sketch and finished it using a mix of AI tools. I also made this as an example of creating free AI art with ChatGPT. 

The prompt grew iteratively. Each unsuccessful generation revealed another ambiguity, so I kept adding more precise constraints until the result came close to the composition in my mind. In that sense, the final prompt is almost an archaeological record of my negotiations with the model.

The image is not only my idealised vision of Touhou Project, but also my own philosophical allegory. The three figures above represent hereditary, ritualised authority and the power of tradition. Sanae, as a modern progressive arriving from the twenty-first century, and Youmu, imagined as a solitary and disciplined swordswoman, confront them from below. Together, the two heroines represent different aspects of the player—essentially, different aspects of myself.

The input prompts are as follows: 

My codes of financial time series analysis and the AI art used in my Yukkuri Commentary movie posted on YouTube

This video presents our recent research on financial volatility forecasting, comparing classical econometric models, signal-processing techniques, and modern machine-learning approaches.

The study evaluates the forecasting performance, volatility prediction accuracy, computational efficiency, and interpretability of five different methods:

• GARCH
• SARIMA
• Matrix Pencil (MP)
• XGBoost
• Long Short-Term Memory (LSTM)

Using Bitcoin returns as a high-volatility financial asset, we investigate whether modern AI-based methods consistently outperform traditional forecasting techniques when accuracy, explainability, and computational cost are considered simultaneously.

Our results show that:

✓ SARIMA achieves the lowest return forecasting error.
✓ Matrix Pencil delivers comparable forecasting accuracy with extremely low computational cost.
✓ XGBoost provides a strong balance between accuracy and efficiency.
✓ LSTM exhibits relatively strong performance in tracking volatility patterns and market regime changes, despite higher forecasting errors and computational requirements.
✓ No single model dominates across all evaluation criteria.

The findings suggest that the Matrix Pencil method remains a competitive and highly interpretable alternative to black-box machine-learning models for financial forecasting applications. 

NicoNico: https://www.nicovideo.jp/watch/sm46580360 

Sakuya, Youmu, and Marisa challenge a unique cooking competition!
Using Bitcoin market data as ingredients, five dishes — GARCH, SARIMA, Matrix Pencil, XGBoost, and LSTM — to compete to predict future market movements ♬
This video explains the strengths and characteristics of traditional econometric models, signal-processing approaches, and modern AI techniques through a fun Yukkuri-style commentary.
Which model will create the best “forecasting recipe” for the future market?

Artwork including original character illustrations:
https://www.pixiv.net/en/tags/%E5%A6%...

Music, sound effects, and background materials used in this video are credited in the end credits.
#TouhouProject #Touhou #YukkuriCommentary #reimu #marisa #sakuya #youmu #remilia #yuyuko #Finance #Bitcoin #Cryptocurrency #QuantitativeFinance #FinancialEngineering #FinancialForecasting #VolatilityForecasting #Econometrics #ARIMA #SARIMA #GARCH #ArtificialIntelligence #MachineLearning #DeepLearning #DataScience #TimeSeriesAnalysis #ExplainableAI #XGBoost #LSTM #SignalProcessing #MatrixPencil #Prony #PronyMethod #Python #Research

This image is posted on my pixiv page: https://www.pixiv.net/en/artworks/147674029

My attempt is to generate an AI art with free of charge using ChatGPT. Many AI illustration tools charge fee and are not so flexible yet. I have experimented with generating an image based on my rough drawing together with the input script into ChatGPT.

This is an AI-synthetic illustration output by ChatGPT based on my rough drawing. The following is the input script. It is like asking the contemporary popular character designer to illustrate Touhou Project characters in his style. ☆彡

Redraw this illustration in a 1990s Japanese fantasy RPG illustration style with the scene shows three cooks working side by side in a cozy kitchen:

• on the left, a refined silver-haired maid in a blue-and-white uniform with a green ribbon, skillfully slicing vegetables with a kitchen knife

• in the center, a white-haired swordswoman in a green outfit, attentively stirring a pot of simmering food on the stove

• on the right, a blonde witch-like girl wearing an oversized black pointed hat and a black-and-white outfit, energetically stir-frying ingredients in a frying pan.

Please improve the line quality, proportions, and shading while preserving the original composition.

 

*  I have got to prepare for a cover image for a video I am currently creating to post YouTube. Furthermore, over here, it is too hot to be concentrated with my usual art works. .... seriously, it's tough enough to make me sleep quite a few hours! In an accommodation where I am currently living, there is no AC! My energy has recently been drained without noticing. Therefore, I just decided to spend my experiment in the AI output while my concentration and energy for my own artwork is sizably limited