I fucking LOVE this spotlight attachment on this light in the studio. You can shove whatever shaped cookie cutter looking stencil in it and it gives you that shape. In this case, I was using two flat pieces to try and form the thinnest sliver of light to shine across the STUNNING Maria S. She cut her hair recently, super short, but every time we shoot she gets in trouble so I'm not sure if her and I are allowed to work together while she looks this good, haha. As you can see, I opened up the light for the last half of the shoot because I *had* to capture this incredible dress that she had on. I'm also really digging these simple flat colored scenes, feels like I'm making a color swatch.
I started digitizing Anais Dior's face a few weeks ago and it has been a journey.
These are all txt2img with not controlnets in place. I wanted to have a pure model without influence of anything other than the prompt.
The first four images were my first attempts using the Lora method, I did about 5 attempts and sent her the best from those groups and she said "they look nice, but they don't look like me" and I was a little offended (because of the amount of work that went in lol), but she was absolutely correct. While it has some elements that are similar to her's, it's off by enough that it doesn't even come close to passing the side-by-side test.
I started doing more research into other SD users who post trained models to Civit and all the bigger ones seem to swear on Textual Inversion (TI) for people's faces vs Lora, which apparently is better for styles. Now this isn't reflected in the NUMBERS of models that have been trained under Lora vs TI, but I'm starting to realize it's probable because of 1. hardware requirements to train a TI and 2. Lora's are just flat out easier to train. So this past friday I started from scratch to train a TI model of Anais and it was a HUGE pain in the dick. I made 6 attempts, with every iterations being almost a complete do-over because the problems seemed rooted in the initial tags that you do for each sample image (1girl, anaisdior, smiling, etc). There are TONS of writeups online from these 'trainers' but they are anecdotal at best as we are all fucking winging it with this tech. Regardless, I ended up with a prettty good model for txt2img that can be improved by using inpaint (where you replace parts that didnt render correctly, like a face or hands or tiddies lol). It also insists on making her eyes blue by default, even tho they are clearly hazel in the training dataset? AI gonna just lie I guess.
Everything from image 5 onward is the latest model, all txt2img with no inpainting or touchups as I want demonstrate how close we can get to "real" people. Is it perfect? Nah. Would it trick someone who perhaps hasn't seen Anais in a minute? Absolutely...maybe. One thing is for sure, for the last couple of images, there is enough "Anais" there characters in other styles that are inspired by her *look*. Disney Anais, Anime Anais, anything is possible with this tech! The last pic, btw, is what happens when the AI doesn't know what a "tron lightcycle" is, so instead it puts the character over a regular bicycle with neon lights in the back lmao.
Anais fans, what do you guys think?
Thanks for letting me nerd out about this stuff here, I don't have anywhere else to do it!
I've largely been messing with generating pictures using AI, but this morning I wanted to take some time and try and lean further into the video process to see what I could come up with. I'll break these down in gallery order.
The reference sources are included and they are from my catalog, I'm sure you guys recognize some. The AI models used (aka "Lora") were trained by me using photos from my catalog. So I feel very strongly that I am using AI in an ethical, artistic manner. I know this is a concern for some folks. In my eyes I'm remixing my stuff with the internet's stuff.
I'm using Stable Diffusion's img2img + ControlNets + EbSynth and Eb Utility (that probably looks like word diarrhea to some)
The method I used doesn't allow for fine tuning of the prompt afaik, so if the prompt says "Beautiful woman with big boobs" and there are moments in the video where boobs are not visible, the AI will find somewhere to put two big boobs. It might be on the person's arm, it might be on their forehead, it's anyone's guess. So my prompt in all these is typically very vague. "Woman, detailed skin, nude" us the gist of what was used for most of these.
In gallery order: #1 - This take was so bad I almost didn't want to include, but I did anyways because I'm gunning for that "most improved" award. Major issues with face tracking when the video is so soft, hence the constant shifting of the face. I'm using the default keyframe length on this, too, which is fairly long and the reason it just looks like a slideshow with morph transitions. Next step: decrease time between keyframes, use a clearer source.
#3 - I kept the denoise factor low for this one and did my best to tune the OpenPose and depthmap controlnets to keep the generated model as adhered to the source as possible. About 2/3rds the way in there's moments where the arm goes nuts, that's because the AI didn't know how to to interpret the shape of Carolina's super thick braids and thought they were arms, I guess. Lowering the denoise also helped keep the background in check here, but starting a scene from a face and zooming out too full body was a LOT of change for the AI to chew on. Solution: Use steadier or more anchored source.
#5 - The source here, despite being in constant motion, is still well anchored on Joy, with the only big movements being her right arm. I tried using "large_breasts" as a keyword in the prompt and quickly realized that each generated frame was casting its own version of big boobs, so the size fluctuates wildly. I also generated and used a mask for each frame, acting as a dynamic cutout to place the generated model directly over the source material. The second video, in the lagoon, is the denoise factor cranked to the point where none of the source is mixed into the generated image, giving you an idea of the wide gamut of possibilities. I also included the prompt "holding a chalice" and it interpreted that as "place challises wherever". Good times.
#8 - This was my final attempt this afternoon, so I upped the resolution a ton which made it take wayyy longer to process. Crystal clear video of JMO, lots of movement but I figured the clarity would help. It did, some, but the backlight was still pretty overpowering and we lose tracking occasionally. Overall it's a massive improvement over my first attempt this morning, so I'll take it. For these renders, I did my best to keep the denoise jussstt tuned enough that we'd keep a chunk of the source material while still generating what looks like a fully rendered video. This meant not being able to do things like "remove window and replace with picture a painting of Bob Ross in the style of Bob Ross" because we were so reliant on the source layer. Using the depth map, it is possible to create an alpha layer 'behind' the model where the AI can generate whatever, but the source video is too flat for that to be effective. Possible next steps: Find even more stable material, and/or try a completely different method. EbSynth, while still one of the best, is just one of several methods of creating AI video, and there's a new method every damn week.
Observations: This process strongly favors video that was recorded on a tripod or a slow gimbal. The steadier or more focused/anchored to a subject your video is, the better the AI's interpretation of the source material will be. As is with images, the quality of the Stable Diffusion model (aka Lora) has huge impacts on the quality of the video. If you don't train your models faces on almost every angle, the AI will have to do more work to interpret what that person's face would look like at that angle, you get me? lol. Lastly, there's another method I've yet to explore that involves putting up to 16 keyframes on one reference sheet, encouraging the AI to interpret all of them simultaneously which would greatly reduce flicker.
Lahanna in panties and a tiny tank top is a formula for a good afternoon if I do say so myself. I have so much content with her and I feel the need to get more because each time we work together she gets exponentially more comfortable with shooting basically whatever. More to follow!
I posted a few pics from the previous scene with Edyn Nyx (check tags) to my insta this morning (@akamikebphoto on there, too) and got a couple DMs from folks looking for more so here's the followup. Edyn brought some insane wardrobe to this Vegas hotel where I had this whole blue and pink theme going and we just shot a bunch of shit all over it. Her big ol' bolt on tiddies and pink tutu were just too good of a look lol.
If you're looking through these images of Lindsay and you're thinking they look a little off, you would be correct. These are AI images that I generated using Stable Diffusion running on my PC. It takes a long time to fine tune a theme like this when using custom models faces and bodies (yes, I trained it on her nudes that I took so these AI boobs are like 75% Lindsay's actual boobs lol), and then to do all the inpainting to try and clean it up, the struggle with the HANDS, but I really like what I'm getting out of it!
If you're savvy, check the last image for the setup
Seems every time Lindsay and I shoot she pushes things just a bitttttt further. 👀 (gonna edit this and add more later, but Im about to have a a call with her about setting up more shoots and AI stuff in like 15 minutes lol)
edit: Just got off a video call with Linz (who was looking hot af tonight btw). I showed off all the AI stuff and we talked about shooting some more saucy content and she's down for all of the above so yyyeeaahhhh!
Got a video follow up for that photoshoot with Bria. One thing I love about shooting using existing lamps and continuous lights vs flash photography is how easy it is to adapt the colors and vibe of the photos to the video. I actually wanted to invest in an 8k video shooting monster and just extract frames from there to make photo sets. It's not much different from people shooting burst a ten thousand fps, and you'll be able to capture scenes in half the damn time because you wont have to shoot photo -> reset the vibe -> shoot video.
lmao I'm gonna have to start tagging things that way with how good some of these AI renders are coming out. Anyways, happy friday ppl, you may recognize both Bria AND this room. I've shot several times in this spot and I've tried to switch it up each time to give each model a unique vibe. I put one of my little RGB panels in the ceiling fixture to add more kick with that green, and the key light is one of those little round "sunset" lights that amazon is always trying to sell me! I was SHOCKED at how bright it was, and the color gradient adds an interesting flair to an already interesting set.. and subject 😬
Yooooooooooo AI shit is crazy! Here's a walkthrough of where I'm at with this new technology and how I want to integrate it into some of the photography works I'll be working on in the future. Watch this video and drop some thoughts in the comments, I'd LOVE to hear what you guys think after you see what I'm able to do with it.
I've got a good story for this one.. Heather had a large clear bandage on her leg that was starting to peel back and we decided to take some scissors to it so there wouldn't be a lot of work in post. Well, I'm down there trying to clip this super thick bandage material with scissors as sharp as a fruit rollup, we're blabbing away about some bullshit, and I fucking scissored, hard too lol. I felt absolutely terrible, and she about shoved me over the bannister, but we managed to press on and deliver a pretty dope set. I apologized profusely, of course, and I'm prrrreeetty sure she's forgiven me.
There's BTS going out for this with some great POV, btw, but I cut out the assault, haha.
I told most of the story behind this shoot in the video post, but I also want you guys to know that I go through a lot of trouble removing plugs, switches, ugly lights, and stuff that y'all never see nor would you care about, haha.
This set is AMAZING! Heather completely owned this bodysuit jfc. We are literally rocking to Nine Inch Nails (one of her favs, as well as mine) while shooting this, then we switched it up and shot video, then back to photos, it was a fast paced shoot that lent its vibe to the outcome. Very pleased with Heather in these.
POV BTS video (preview in gallery) available as well, tip $10 here and I'll dm you the vid asap!
I've got a few Lahanna things in my queue, as you can see, and this video is dropping on the TL because it's SUCH a tease of the photoset. She doesn't do nudity in videos,or BTS with any photographer it seems, and that's kind of a bummer but I'm hoping the more we shoot, the more comfortable she is with trusting me with this kind of work *because Im fucking good at it*
SOONTM
I know I'm a little behind on my inbox, I really did let chatgpt and scheduling run the show for a minute there so I could work on other projects lmao. I'll get to you guys asap!
Lahanna as you've literally never seen her before.. with her adorably nerdy glasses and braces on full display. She's very protective of how she's perceived on the internet, and this is certainly outside of her comfort zone, but she loved it anyways. Can we talk about the transformation from first to last image?
Probably a Mike B's top 3 shoots with my girl Heather here. A+ Hair/Makeup, A+ setting, A+ wardrobe, A+ lighting, A+ Heather, DDD+ boobas. The VIDEO for this is incredible as well, and the BTS is.. saucy👀
Carolina G is indeed a G for this set. Fully nude in the corner of this airbnb wearing the infamous angel wings (or wing, as it were). Been poking at this one for a while until I one day decided to straight up recolor the entire scene from blue to teal and now I love it!
Bria aka @ yuge.nn with her yuge collection of of vinyl. We spent way too much time flipping through records instead of shooting, but I loved that she stopped on Cheech and freaking Chong haha.
Slithering in her stunning, sultry snakeskin ensemble, she sashayed with spectacular sass and style.
lol
What do you guys think of JMO's new nip piercings? She said they were HEAVY but she liked how it felt, so now she has me wondering if she'd be interested in shooting anything with nipple clamps 🤔🤔🤔
I woke up early as fuck for this, but it was totally worth it. Lindsay is and always will be a playful one, especially when naked. Although lately she's been giving us *ahem* _more_. We had a 3 minute window where the sun would push through and give us these beautiful god (goddess?) rays and we maximized every second of it.
I missed out on working with Jordan West recently, she had a last second emergency she had to attend to and ended up cancelling. It's a shame too because this is the last set I have from our previous shoot! I know there's a lot of fans here, tho, so know that I'm doing my best to get back out her way so we can link up again and get some content done.
JMO looks next-door-gorgeous in this hammock surrounded by.. an actual bamboo forest. The top she's wearing was knit by a friend of hers, so she felt like one classy tiddy reveal wouldn't bring TOO much dishonor haha. I found these prerolls that look like cigs, gonna stock up and try not to smoke them as they're perfect for shoots where we want to get smoking shots but not use actual cigarettes.
Heather and I were prepping to shoot a scene when she steps in wearing this elegant and beautiful dress. This is just a little bit of sun (and a moon..) with a dresser and painting as a backdrop, shot it on my 85mm so my back was REALLY against the wall here. Was a quick set tho, I just wanted to capture this moment.
Shot with Bria in this insane green-room style living room. The owners literally daily drive this space, it's wild. Biggest challenge of the shoot = keeping the camera from autofocusing on Bowie's face instead of Bria's lol. I've also included a shot of the lighting rig I used, as it's become sort of a thing since shooting that black and white duet set with Linz and Heather a couple years back. I really appreciate the extra layers of colorful shadows it creates. Dropping the video on the TL tomorrow!
How timely to post my non-nude content on a Sunday yeah? 🌩😬🌩This set with Maria was meant to have a bit of an edge to it but, once the scene was lit, I told her it looked like a gdamned LinkedIn portrait shoot.. but we should press on because it didn't technically look BAD. What she's wearing is stunning, ofc, as you can see every curve of her hella fit body through this tight knitted romper. I laughed when we got to "stage 2" of the shoot where normally clothes would start coming off and we undid like two buttons haha.
Here's Heather rocking the Sunset backdrop scene! We were racing the sun so we shot this in literally THREE minutes, I have the entire POV BTS going out via DM for this interested. Heather's great for getting in and out of scenes quick and still getting some gangster shots, an absolute pro. This jean jacket sure does get around tho 🤔
