Gemini getting better and better in generating images: Image Consistency, and Spelling

 I still remember not long ago, Gemini would sometimes display 2 problems when generating images: (1) image consistency of the main characters in a picture sequence, and (2) spelling. 

Today, those issues seldom occur again. Good job, Gemini.

Repicating someone's created comic strip and turning the dialogues into English: Good Results

Original Comic Strip by Ms Selena Tam: 

WHAT I DID: 

1. I went to Gemini > Image.

2. Paste in the above comic strip. 

3. Wrote the prompt: 

Replicate the uploaded comic strip, turning the Chinese dialogs in the speech bubbles into English

4. output below: 

================================

TAKEWAY: 

1. Gemini replicated the original images, without changing them in any way. 

2. The English translation from the Chinese is accurate. 

3. If teachers find a suitable comic strip in Chinese which they want to use in an English lesson, they can follow this example.  

 

Created a 3x3 choice board using Gemini's Create Image function: Very straightforward

 In Gemini's Create Image function: 

MY PROMPT: 

Create a 3x3 choice board. Each of the 9 boxes contains (1) the image of a teenager hobby (e.g., playing games), and (2) a caption (e.g., Playing Games). The choice board should be a square, not a rectangle.

OUTPUT:

NOTE:

Gemini's default image size is a rectangle. Need to specify that a square format is wanted.

Gemini's "Create Image' function: PROJECT AC vs PERSONAL AC

1ST TRY

PROMPT: Create an image that illustrates Newton's first law of motion 

=====================================================

PROJECT AC OUTPUT: 

PERSONAL AC OUTPUT

=======================================================

2ND TRY

PROMPT: An image that explains how a typhoon is formed. 

PROJECT OUTPUT: 

PERSONAL AC OUTPUT: 

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

OBSERVATIONS

1. Outputs can be different, eg, the number of stages. 

2. No obvious 'better' tool. But Personal AC Output might look slightly 'better.' 

Canva.com to edit text in Image: List of useful functions

 Earlier on, I created a poster using Gemini. The visual output is fine, but there was a spelling mistake. 

Thanks to Ada Lam, who alerted me to the "EXIT TEXT IN IMAGE" function in Canva: 

1. 

https://www.canva.com/features/edit-text-in-image/

2. Upload problem image. 

3. Grab text box in uploaded image. 

3. Click 'Edit' icon: 

4. Choose 'Fix Spelling'

5. Check 'Replaced Text'; click ''Replace" if accepted: 

===========================================

Tried the Infographic function in Piktochart AI: Results disappointing

 1. I uploaded a pdf file. 

2. Chose the 'Infographic' function. 

3. In context, asked for 'a structured summary of the uploaded file'.

====

RESULT

- not a structure infographic. 

- more like a poster.

- no structure. 

- visuals are purely for docoration only. 

Created a comic strip from a plot using Nano Banana

APP: Nano Banana in POE 

My PROMPT: 

Please create a 4-panel comic strip. Here is the story: 

- A grandfather and his 7-year old grandson visited a pet shop which sells hamsters. 

- Grandfather asked Grandson whether he knew how to draw hamsters. 

- Later, Grandfather received a drawing from Grandson. It was the drawing of a hamster. 

- Grandfather was very pleased with the drawing. 

WHAT HAPPENED NEXT: 

Nano Banana, while 'thinking', enriched my expectations with further details. 

It then produced 3 (4?) versions of the comic-strip, WITH DIALOGUE. 

There was a grammar mistake with one sentence in the dialogue. 

OBSERVATIONS

- Image consistency is no longer an issue. 

- I don't need to provide it with the dialogue. 

Turned an infographic into a text-only graphic organizer using Gemini

 Infographic generated from previous attempt: 

In Gemini, I wanted to convert it to a text-only graphic organizer which I might turn into a blank-filling task later. 

My first prompt: 

Please convert the above image into a black-and-white graphic organizer chart with text only, i.e., without the pictorial details. 

Output after 2 revisions: 

The first output was not exactly what I wanted. I then attached the following image to my prompt to clarify what I needed: 

and rephrased my prompt: 

Please follow the attached graphic organizer example, and turn the above chart into a text-only chart. There should be no images in the chart. 

OUTPUT: 

So, I finally got what I wanted. 

Created a 'Layers of a Forest' graphic organizer using Gemini's 'create image' function

 MY PROMPT AND INPUT

- a reading passage on the layers of a forest, in image file format. (I tried uploading different file formats, such as PDF, DOCX, but only image files work.

- My prompt: 

"Use the content of the attached information to create a graphic organizer showing the layered structure of a forest."

OUTPUT: 

OBSERVATIONS

1. The output is actually like an infographic, not a graphic organizer diagram. 

2. It DOES make use of my input information, though I could have used a topic as my prompt. 

3. Gemini has done quite a good job in generating this image. 

Created a picture using Gemini-2.5-Flash-Image and Ha Sir's prompt

 TOOL: Poe > Gemini-2.5-Flash-Image

PROMPT: (provided by Ha Sir)

Create a 1/7 scale commercialized figure of the attached photo, in a realistic style and environment. Place the figure on clear acrylic stand. Next to the computer screen, display the ZBrush modeling process of the figure. Next to the computer screen, place a BANDAl-style toy packaging box with title “XXXXX“ printed with the original artwork.

OUTPUT