Comments

Blacklisted
    (help)
    Show Search Options
    post #168954
    21 20 3 S
    post #169923
    11 1 2 E
    post #167720
    179 307 5 E

    LionXLD

    Member

    tempting said:
    The base gen started with just the guy and lion bumping uglies, everyone else was inpainted in. When it comes to groups, it's exponentially easier to place the pieces yourself instead of hoping regional prompt reads your mind (which for me is almost never).

    So you used Photoshop to stitch them together in post-production? Your work is so perfect, I can't see any trace of stitching at all.

  • |
  • 0
    •
    post #169952
    25 36 2 E
    post #169956
    7 8 1 S

    n n

    Member

    ❤️ Lovely calm moment in the summer grass. This is wha ti should be doing rn

  • |
  • 0
    •
    post #169952
    25 36 2 E
    post #169923
    11 1 2 E
    post #169932
    26 39 1 E
    post #169511
    19 9 6 E
    post #169851
    58 53 7 E

    mlem

    Admin

    Very cool project, Nilmi, thank you for sharing with the rest of us!

    I've loved hunting for the lil easter eggs too, that was a really fun idea! The one on the second page is giving me trouble though 👀 Time to lock in.

  • |
  • 2
    •
    post #168679
    285 410 10 E
    post #169863
    47 59 2 E
    post #156643
    125 203 1 E
    post #119465
    94 128 2 Q
    post #169937
    74 96 1 E
    post #169921
    18 12 7 S

    n n

    Member

    bahufaru said:
    As I work with Google Gemini, I have reached a point where I can feed it multiple reference images and detailed descriptions, have it ingest them all and produce a pretty good model, and then send that to the animation pipeline, including dialogue lines and directions for voice types and accents.

    The prompting for this wasn't particularly difficult, and it's in natural language, basically just "Animate Kafeel Scirocco chopping vegetables in a school cafeteria kitchen, in a photorealistic style similar to student video productions. Have him regard the viewer and say, in his Sudanese accent, "My name is Kafeel Scirocco. It is my jobs to nourishes the hungries.""

    The real work is in the significant amount of reference images and concept text, character explaining, etc, that went into reaching a point where Gemini fully understood the on-model design of the character and could fully understand what "Kafeel Scirocco" means, conceptually and visually.

    The nice thing is that once that work is done, my instance of Gemini reliably produces media of the character, because it understands what the character is and what he should look and sound like.

    It's very impressive :D and thank you for explaining ❤️

  • |
  • 0
    •
    post #169921
    18 12 7 S
    post #168413
    52 59 2 E
    post #168919
    90 130 4 Q

    Chamir136

    Member

    corsetlover said:
    Would love to see more of the laughing cow from you! Shes a great character and i have a soft spot for bovines. :)

    Thanks and i do have a few more pieces with her to post so dw, if you have a suggestion on gens with her feel free to say

  • |
  • 0
    •
    post #169130
    102 133 1 Q

    Corsetlover

    Member

    I’m shocked nobody has commented yet! I love everything about this piece. The angle is great as well as her pose and expression. Great job!

  • |
  • 0
    •
    post #167110
    61 81 1 S
    post #166780
    141 191 3 Q
    post #145207
    48 76 1 Q
    post #166302
    95 131 1 Q
    post #166185
    59 75 1 S
    post #165886
    142 167 1 Q
    post #168919
    90 130 4 Q

    Corsetlover

    Member

    chamir136 said:
    Thanks a lot, glad you like it !

    Would love to see more of the laughing cow from you! Shes a great character and i have a soft spot for bovines. :)

  • |
  • 0
    •
    post #168793
    39 63 2 Q
    post #168933
    47 76 1 Q
    post #169921
    18 12 7 S

    Ocillet

    Member

    bahufaru said:
    As I work with Google Gemini, I have reached a point where I can feed it multiple reference images and detailed descriptions, have it ingest them all and produce a pretty good model, and then send that to the animation pipeline, including dialogue lines and directions for voice types and accents.

    The nice thing is that once that work is done, my instance of Gemini reliably produces media of the character, because it understands what the character is and what he should look and sound like.

    Damn, that's interesting. I haven't played around with the larger public models since the early Kling days (and nowadays I just gen everything locally to not deal with NSFW filters), but that's pretty amazing you can just feed Gemini information and references to build a LORA of sorts and retain a character for different scenes.

  • |
  • 0
    •
    post #169921
    18 12 7 S
    post #169933
    27 37 1 E
    post #169921
    18 12 7 S

    Bahufaru

    Member

    n_n said:
    God damn that's some impressive interactions 😮
    HOW did you manage this??? it's quality is something i would only expect out of a movie

    As I work with Google Gemini, I have reached a point where I can feed it multiple reference images and detailed descriptions, have it ingest them all and produce a pretty good model, and then send that to the animation pipeline, including dialogue lines and directions for voice types and accents.

    The prompting for this wasn't particularly difficult, and it's in natural language, basically just "Animate Kafeel Scirocco chopping vegetables in a school cafeteria kitchen, in a photorealistic style similar to student video productions. Have him regard the viewer and say, in his Sudanese accent, "My name is Kafeel Scirocco. It is my jobs to nourishes the hungries.""

    The real work is in the significant amount of reference images and concept text, character explaining, etc, that went into reaching a point where Gemini fully understood the on-model design of the character and could fully understand what "Kafeel Scirocco" means, conceptually and visually.

    The nice thing is that once that work is done, my instance of Gemini reliably produces media of the character, because it understands what the character is and what he should look and sound like.

  • |
  • 2
    •
    post #169476
    74 108 3 E

    medicineball

    Member

    Once in a great while you find a really attractive monkey girl like this. No surprise to me that it's one of Saggy's creations.

    Perfect breasts, great figure, matching hair highlights to nail polish and matching jewelry...
    But that bright, smiling face has to be her best feature. This girl looks like fun.

  • |
  • 0
    •
    post #169911
    49 78 2 E
    post #169773
    117 128 8 E

    medicineball

    Member

    I'm not attracted to nagas, but this just kills. I am so entertained by the line, the description and this discourse. Thanks Terra and participants, you made me laugh. You made my night.

  • |
  • 1
    •
    post #169921
    18 12 7 S
    post #169911
    49 78 2 E
    post #169921
    18 12 7 S

    n n

    Member

    God damn that's some impressive interactions 😮
    HOW did you manage this??? it's quality is something i would only expect out of a movie

  • |
  • 4
    •
    post #169888
    31 37 2 E
    post #156053
    390 656 10 E

    TailUpSD

    Member

    smallzballz said:
    Love this! Did you gen the story as well? What did you use?

    I write all of my stories completely by hand without any assistance from generative AI. Thank you for asking!

  • |
  • 0
    •
    post #169913
    40 61 1 E
    post #167720
    179 307 5 E

    Tempting

    Member

    lionxld said:
    How did you generate so many characters[...]?

    The base gen started with just the guy and lion bumping uglies, everyone else was inpainted in. When it comes to groups, it's exponentially easier to place the pieces yourself instead of hoping regional prompt reads your mind (which for me is almost never).

  • |
  • 0
    •
    post #169892
    32 52 3 E
    post #155106
    139 199 1 E
    post #152560
    304 478 7 E
    post #138972
    103 150 2 E
    post #138971
    156 213 1 Q

    n n

    Member

    Super serene gen with such a lovely fox. The kind of water you have here that almost looks like a jelly layer is working amazingly well to separate the levels

  • |
  • 0
    •
    post #169920
    59 82 1 E
    post #169905
    42 42 1 E
    post #169908
    78 129 3 E

    ZhenEric

    Member

    Magnificent~ her cookie being a bit more detailed and the way it peeks through her fur is so nice

  • |
  • 1
    •
    post #169844
    38 48 1 E
    post #169852
    37 54 1 E
    post #169767
    124 179 9 E

    wolfucker666

    Member

    Skeleton slimes? Yes please. Though I feel like we're gonna have to be a bit forgiving on the anatomy on these because skeletons are very difficult for AI currently

  • |
  • 0
    •
    post #169908
    78 129 3 E
    post #169720
    41 54 3 Q

    n n

    Member

    denatural said:
    It's true! the tags says so!
    Poor guy, spent from too much sex.

    Even big strong dergs like to get dom'd :P

  • |
  • 0
    •
    post #119633
    81 135 1 Q

    gridanon

    Member

    This was a fun one to do that came from experimenting with different clothing on characters. The bottom here is called a "sarong" and it'd be neat to see more unique clothing like that in gens.

  • |
  • 0
    •
    post #169908
    78 129 3 E
    post #169831
    16 26 2 E
    post #123262
    66 80 5 S
    post #169275
    45 55 1 S
    post #164952
    44 76 1 E
    post #151870
    106 176 4 Q
    post #169899
    17 19 1 S
    post #169894
    35 40 2 S
    post #169888
    31 37 2 E
    post #169882
    108 169 2 E
    post #167120
    129 195 2 E
    post #151581
    217 375 2 E
    post #122970
    203 344 3 E
    post #169864
    45 60 2 Q
    post #40444
    148 256 2 E
    post #169857
    64 76 3 S
    post #169851
    58 53 7 E

    n n

    Member

    This has been a Wonderful cruise with tons of excitement and an absolute boatload of amazing characters.
    i give it 3 thumbs up and a standing ovation *Applause*

  • |
  • 2
    •
    post #169894
    35 40 2 S