Free AI web copilot to create summaries, insights and extended knowledge, download it at here

Abstract

src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*ueMM86ELejTaukxy9cbuVQ.png"><figcaption></figcaption></figure><p id="cd34">BTW, the scrolling results on the Discord UI are annoying. We can open the Inbox in the upper right corner and check the post mentioned to us.</p><figure id="5637"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*Ci4No32HYhivx1_ELbPz9g.png"><figcaption></figcaption></figure><h1 id="058c">2. How to create the same or similar output every time</h1><p id="3d89">I don’t think the same prompt leads to the same images each time. I guess it could not be possible because of the trait of a Generative Adversarial Network(GAN), but it is worth listing here. Because the ability to reproduce the same image with the same prompt plus some initial parameters could be powerful.</p><figure id="e202"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*E7sxIo337N7iesqL04wa6g.png"><figcaption></figcaption></figure><p id="222b">The one I created with the same prompt</p><figure id="93b2"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*d69NL_k7EgSxJl9vRoUH9Q.png"><figcaption></figcaption></figure><h1 id="7788">3. How to draw specifically</h1><p id="4e7e">For example, I want to draw my avatar, Moogle, from an online game FFXIV. But MidJourney doesn’t know it; what can I do?</p><figure id="d191"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*bBwg-Z4g5e9CuLnFBnAgkg.png"><figcaption>I don’t want this mouse.</figcaption></figure><figure id="218a"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/0*xw6JBhsxxByOG7Cf.jpg"><figcaption>The Moogle</figcaption></figure><p id="6713">It’s worth mentioning although MidJourney doesn’t know the difference between Inkling and Octoling in Splatoon, it “understands” the Splatoon style, the bright, colourful style. I love it.</p><figure id="7e51"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*mxSzNZzdLrJc2h1sOLzxYA.png"><figcaption></figcaption></figure><figure id="2abf"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*bW9Cze6JEf4vj35yIl8TiQ.png"><figcaption>upscaled</figcaption></figure><h1 id="ab38">4. The Blend function</h1><p id="58b9">I tried /blend and uploaded two pictures because I want the original Moogle with Nier Style. This attempts to solve the las

Options

t problem, but I can’t get the figure from the picture.</p><p id="16fc">Uploaded picture one</p><figure id="857a"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*QhblJfDQMU3v1yplcEAXyg.png"><figcaption></figcaption></figure><p id="5d57">Uploaded picture two</p><figure id="fe08"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/0*RChRCX6kOPZvUW_y.jpg"><figcaption></figcaption></figure><p id="537d">What I got</p><figure id="30cb"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*--aDPUbgL1tlJfJR3uAUWA.png"><figcaption></figcaption></figure><p id="978c">I also tried just uploading one image and asking it to make it black and white; the result is meaningless.</p><figure id="5890"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*UKHaaxI0w_E7A718ymZ-1g.png"><figcaption>Namazu original picture</figcaption></figure><figure id="790f"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*jFnZXzLkO_t6kHMXc2YuXA.png"><figcaption>The Namazu in the last picture turn to something strange</figcaption></figure><p id="30c8">To summarize,<b> I’d like to draw exactly what I want; some meaningful characters already exist but are not that commonly mentioned. </b>I will study how people use MidJourney and see if I can improve my drawing and make them more specific.</p><p id="6a33">If you found it helpful, you could read these posts about using AI tools to improve our productivity. Thanks.</p><ol><li><a href="https://reneelin2019.medium.com/safe-ways-to-use-chatgpt-to-study-better-no-cheating-c1712d61c123">Safe Ways to Use ChatGPT to Study Better (No Cheating)</a></li><li><a href="https://readmedium.com/three-ways-to-upload-pdf-to-chatgpt-to-analyze-or-summarize-59ffb059f7e3">Three Ways to Upload PDF to ChatGPT to Analyze or Summarize</a></li><li><a href="https://reneelin2019.medium.com/how-to-use-dall-e-2-to-create-images-from-text-17e041f81614">How to use DALL-E 2 to create images from text</a></li><li><a href="https://reneelin2019.medium.com/how-to-use-stable-diffusion-to-create-images-from-text-bbbc8a419f9">How to use Stable-Diffusion to create images from text</a></li><li><a href="https://reneelin2019.medium.com/how-to-use-dreamfusion-to-generate-3d-models-from-text-d5eb6d37a1c7">Using DreamFusion to Generate 3D Models from Text</a></li></ol></article></body>

Four Problems Bother Me When Using MidJourney+ChatGPT as a Beginner

https://www.allabtai.com/chatgpt-midjourney-v4/

Recently AI Tools kept evolving and catching greater attention compared to the time DALL-E 2 was released last year. Like most people, I wonder how I can use those tools. ChatGPT has been helping me in my study (Safe Ways to Use ChatGPT to Study Better). This urges me to explore how to use AI art tools since I have always wanted to improve my drawing skills.

Besides testing how to use ChatGPT with MidJourney together, the same problems bother me as I tried DALL-E and Stable Diffusion six months ago. (I will stick to MidJourney as I think it is the best)

I’d like to summarize the problems here and solve them eventually in the near future.

How to use ChatGPT together with MidJourney
How to create the same or similar output every time
How to draw specifically (e.g. A manga character fewer people know)
I tried the Blend function and felt it was more difficult to master

Can I use ChatGPT to improve the current images? Especially for the problem 3, 4

1. How to use ChatGPT together with MidJourney

I tested if ChatGPT can draw me a Pikachu, but it’s more like an EEVEE.

BTW, the scrolling results on the Discord UI are annoying. We can open the Inbox in the upper right corner and check the post mentioned to us.

2. How to create the same or similar output every time

I don’t think the same prompt leads to the same images each time. I guess it could not be possible because of the trait of a Generative Adversarial Network(GAN), but it is worth listing here. Because the ability to reproduce the same image with the same prompt plus some initial parameters could be powerful.

The one I created with the same prompt

3. How to draw specifically

For example, I want to draw my avatar, Moogle, from an online game FFXIV. But MidJourney doesn’t know it; what can I do?

It’s worth mentioning although MidJourney doesn’t know the difference between Inkling and Octoling in Splatoon, it “understands” the Splatoon style, the bright, colourful style. I love it.

4. The Blend function

I tried /blend and uploaded two pictures because I want the original Moogle with Nier Style. This attempts to solve the last problem, but I can’t get the figure from the picture.

Uploaded picture one

Uploaded picture two

What I got

I also tried just uploading one image and asking it to make it black and white; the result is meaningless.

The Namazu in the last picture turn to something strange

To summarize, I’d like to draw exactly what I want; some meaningful characters already exist but are not that commonly mentioned. I will study how people use MidJourney and see if I can improve my drawing and make them more specific.

If you found it helpful, you could read these posts about using AI tools to improve our productivity. Thanks.