Hi all, Recently, I've been trying to get better at drawing. It started with generating AI scenes and editing them, but as my desire to create more complicated scenes and interactions grows, I wanted to invest in my own hand skills to achieve (And to develop my own style). It's come to a point where I feel comfortable posting images that I've drawn (Meaning no AI used directly or only used for background). The story in question is here: https://chyoa.com/story/My-Mom-is-a-Maternal-Slime.45566 WARNING: I really picked the weirdest stuff to publish for the first time! Mary-Anne Conjointment Splitting: Here, we have two characters from the story, Mary and Anne. Anne is a slime girl and currently is mimicing Mary's appearance Camera Screen Overlay: Without camera screen overlay: No, they're not actually physically conjoined. Rather, Mary is "wearing" Anne. They're in the middle of a shot to create an interesting photo reference that portrays "conflict", for which Mary could use for a future painting idea. Speaking of reference, that's exactly what I used. Here is my first attempt... To learn how to improve, I input this into the AI to see how it would change it. The result is the reference picture on the side. After that, I learned that the nose bridge should be sharper, mouth needed depth, and eyes should be lower (In line with nose bridge). I attempted to redraw and fix those issues, then separate them with stretch lines. Mary-Anne Fusion (Second Skin) So, this is the second drawing I did for the same chapter. The wood floor/walls is AI generated (Which I plan to replace later as I'm figuring out how to best use Krita to draw parallel/even lines in an efficient manner), but the main focus is drawn by me. I'm pretty proud considering I spent a lot of time trying to figure out where best to highlight and shade. Hands and feet are, of course, hard to get for a beginner. I'm still figuring out what parts of the feet need to be larger/emphasized to look right. Her hand on her breast is okay, but I did not want to mess around getting into a gripping form.
Currently, I'm rewriting a chapter where Anne sucks off Daniel in the first sexual act she performs for him. As part of that, I'm also drawing her in a more goopy form. The goal is to eventually replace the AI images of Anne in her slime form first. I'm still playing around with her look with a few other scenes, but so far, I'm satisfied with this image which I plan to use: Update #1 (Depth was missing from the slime's back, so I've edited it accordingly): Update #2: Added back hair. She has longer hair, so it doesn't make sense that it would slide to the front and disappear. For comparison, here's the AI image that'll be replaced as part of the chapter rewrite:
I'm not sure. Originally, I did grayscale to keep generations consistent and to make editing easier. It's different now where I have the layers set up and can recolor individual parts as needed without too much effort. That said, there's something to be said about the simplicity. I don't know, maybe I'll color/emphasize special moments.
More drawings! Above, I've added more goop hair and fixed Daniel's hands on the bed. In addition, I've made an sticky progression in the rewrite that I ultimately scrapped. I didn't want too many images in a single chapter anyways. Just for fun, since I do have similar images, I tried Google's FiLM interpolation to see if it could create an animation: I'm actually surprised! The messy transition could be fixed with an intermediate state and some edits, but considering I wasn't planning any animation, it generated transition frames pretty well! I can see how a person could use FiLM to generate the frames to assist them. Anyways, leading up to that scene, I have three attempted drawings: Attempt #1: I was going to use this, but ultimately decided to scrap it. While cute, the characters look younger as a result. Plus, I didn't like how Anne was sitting on on the bed. Of course, I could've fixed that by experimenting with shading, but ultimately, I decided to redraw. Attempt #2: Now drawing an easier angle, focusing less on the bed and more on their act. That works out best since I'm making more of the resolution. I mostly stick with this, but I didn't like how stiff Anne's goopy hair looked. Plus, even though Anne looks more mature, I prefer the sharper nose bridge that I drew Mary within the conjoinment picture. As such... Attempt #3: Finally, we have more flowing goop-hair for Anne along with a sharper nose-bridge, which I think looks better. I also simplify the shading and highlights, trying to keep it clean. Here's the AI comparison: Skip a few days, Anne and Daniel have got more comfortable with each other. We see that in this drawing that I decided to color rather than start out greyscale (Ultimately, the image embedded in chapter is still a greyscale version): There's more to the image of course, but cropping it, it looks better with Anne's eyes out of the view. More mysterious, maybe? Another reason is that I'm not happy with Anne's eyes, so I'll have to work on that more. I'll post a more full version once I get that right. However, I liked this enough that I made it the new cover photo:
Off the bat in the introduction chapter, we see a group of robed and hooded members circling a fire and speaking a strange language. When I started AI, this was the image I generated: Now, I've drawn the replacement: Then, we have follow-ups to the terrifying image of Anne becoming a second skin to Mary in the first post: From Mary's appearance, this new fusion now changes herself to a mix between Anne and Mary. She's delighted to meet Daniel with her new identity: Marianne. Revealing herself to Daniel, he's naturally shocked and confused. She's been cooking while wearing nothing but an apron! (Yes, that classic trope).
Melissa: Here we have Melissa, waking up from her unconscious state only to see a towering slime girl (Who had possessed her earlier). I need to work on the head more, but overall she's fit, but not toned like the AI image (Which I love, but I'd rather keep her images more consistent) AI image: Sarah Knight: Sarah, is the new main character for a Non-Incest/Non-Pseudo-Incest branch. Here, I decided to try more simple shading as opposed to the airbrush I've done so far. The good thing about this drawing is that she looks more mature as oppose to the generated AI. Otherwise, I like the simplicity of the AI image which has almost no shading at all. I'll have to try taking another pass on all the characters so far with cleaner/minimal shading. Edit (12/27/22): Reinputing into AI for feedback, I think I've discovered in general what I'm doing wrong. For a front-facing character in manga/anime style, the outline of the head is usually exposed rather than hidden by the hair/locks. In addition, I've distributed bangs out more. While majority is still in the center, it looks more natural and better as a result: Original AI image:
New image. Continuing on with Mary and Anne's fusion, we now see them splitting apart: So I'm generally happy how the heads turned out, but below the neck, I feel like something is missing in terms of shading. Like the above picture of Sarah, I decided to keep the shading simpler without relying on the airbrush tool. Why I'd like to use flat colors is because it makes fixing errors easier. If I have to redo a shadow with airbrush, I usually have to erase the whole shadow part because it's hard to match the subtle gradients that are produced. I don't know, but maybe it's worth it since it adds a sense of depth. Done well and you get a really nice image like the blue slime picture above. Perhaps I'll do a hybrid approach on the image.
Hello! Drawn another image of Marianne enjoying her home-made vegetable stew. Once more, it is simple shading rather than airbrush. I think I've more or less found how I want to emphasize depth/details. And an alt without the apron and table:
Another image and another split mid pounding! I'll be honest, I got lazy with the feet and a little with the hands. This is my first time with the position and further complicated with the slime split mid-pleasure. Definitely going have to work on this one more.
Hi Elmer, I love your work, and the fact you are willing to share your talent with us. A very talented artist. Keep up the work my friend. Keep me posted and I will follow.
We have Anne focused on drawing a strange symbol that she recalls from Mary's memories when they fused. It comes from Mary's uncle, Marlin, but why? (I'm actually not sure about the symbol. I may change that up, but it's minor compared to the idea that something from Mary's family is strange.) Also, I played with animation feature in Krita. I used it as a nice ending animation at the end of the latest chapter considering Daniel is "sinking".
I've struggled so much with Melissa's appearance, but I think I've got two versions that I'm now satisfied with (Or where I'm more likely to modify than to restart from scratch). To recap: This was the original AI image: Yes, with AI, I can't really compete. This *is* the better image; however, I want to have a "hand" in the images I'm embedding to my work. Right now, AI (or specifically NovelAI) might be good for solo pictures, but it really struggles with compositions and multiple people. Then, my first attempt to recreate it: I've since decided to not use airbrushing as the way to emphasize depth and shadows; however, I wasn't really satisfied with the general head shape and eyes. Version 1 (Close-up version with alternate bangs): I do like this, and I'm uncertain between this and below, but it does break away from the general pose of her pushing off against the couch as she wakes up with shock. Version 2 (Going back to slight tilt of body): Definitely less detailed than the close-up above, but you get the original angled pose. I kept the split bangs the same as the original, but added a more concrete shadow of Slime-Amy towering over her. I could copy the shadow and add it to version 1 too. If anyone has their opinion, please let me know.
I haven't released a chapter with this yet, but wanted to share another animation I've finished working on: What I did is draw 12 frames. That's easy enough with only the water and the slight throw of the hair. Then, I tried using Google's FiLM interpolation to generate in-between frames. Really impressed by it and it helps to make existing animation smoother. For comparison, the original 12-frame animation was this:
Another animation, a follow-up from above (She gets surprised by MC and fumbles off the boulder): I didn't do any interpolation for this one, mainly because there's too many moving parts and more frames I'd have to tediously pass through FiLM (At least until I write a modified script, which I plan to for future animations.)
New drawing to replace previous AI image in chapter 7: In the chapter, it's greyscale to keep consistent with the rest of the images. ...and for comparison, here's the previous image: Other note, I've finally set up a DeviantArt page. Of course, I can't post everything there on that, so I'll have to find another suitable place for the more explicit pieces (After I fix glaring errors, that is.)
