<p>You'd assume that a LORA that just shows a pretty face rocking forward and back would be easy! Apparently not. Occlusion is a b!tch.</p><h1 id="front-doggystyle"><span style="color:rgb(64, 192, 87)"><em>Front Doggystyle</em></span></h1><p></p><p><span style="color:rgb(250, 176, 5)"><strong>WAN22</strong></span></p><pre><code>A {slim|fit|athletic|curvy|heavyset|cute|skinny|petite} {Russian|French|Swedish|Swiss|Latina|Austrian|German|Dutch|English|Irish|Portuguese|Indian|Russian|Swiss|Swedish|Danish|Italian} woman is bent forward facing the front, a {dark-skinned||} man is partially visible behind her thrusting into her, he is holding her {hair|throat|mouth},

Her {blonde|brown|dirty blonde|light blonde|bleach blonde|black|redhead|dyed|punk|styled} hair {falls loosely around her face|is tied back|is messy and falls forward around her face}.

She is naked {with {huge|large|medium-sized|small} breasts|, her chest resting on the surface beneath her}.

{She is visibly moaning from pleasure|Her mouth is open and her brow is furrowed|There's string of white fluid around her lips and on her forehead|She is super happy and excited, really into it}.

{bright|soft|natural|indoor} lighting, {high|low}-angle view</code></pre><p>HunyuanVideo</p><pre><code>A{ slim| fit|n athletic| curvy| heavyset| cute} {Russian|French|Swedish|Swiss|Latina|Austrian|German|Dutch|English|Irish|Portuguese|Indian|Russian|Swiss|Swedish|Danish|Italian|Asian|Japanese} woman remains bent forward from the front with a man kneeling low centered behind her, thrusting steadily.

{eye-level face prominent|slightly below eye-level|slight high angle|high angle  showing her butt crack}

Her {small breasts jiggle beneath her|medium-sized breasts swing with each thrust|large breasts swing beneath her}.

{She is very happy|Her mouth is open and her brow is furrowed|She is screaming from pleasure}.

{His right hand is holding her hair|His hand is holding her bangs|His right hand is holding her ponytail tightly on the top right of her head|He is holding her shoulders firmly}.

They are both naked with the scene lit {brightly|softly|dimly with a spotlight on the subjects}.</code></pre><p></p><h3 id="versions"><span style="color:rgb(190, 75, 219)">Versions</span></h3><h3 id="three.1"><span style="color:rgb(253, 126, 20)">three.1</span></h3><p><span style="color:rgb(134, 142, 150)">Same as below but a tiny bit cleaner and slightly better(faster) motion.</span></p><h3 id="three"><span style="color:rgb(253, 126, 20)">three</span></h3><p><span style="color:rgb(134, 142, 150)">Blurred faces and retrained with a select set, pruning most deviations from the core pose. This allows for stability through generalization. 10 videos, 24fps, 48 frames, ~256x256 LR 5e-5</span></p><h3 id="two"><span style="color:rgb(21, 170, 191)">two</span></h3><p>19 videos, 24fps for 50 frames, better captioning and better diversity. We don't need the pillow anymore as the normal resolution run caught the right details down there, most of the time. This one is head and shoulders above <span style="color:rgb(255, 255, 255)">pillow edition </span>but far from perfect. Alas, it's worth the release!</p><p></p><h3 id="pillow-edition"><span style="color:rgb(255, 255, 255)">pillow edition</span></h3><p>This is just a proof of concept I ran. The curve looked great till 140, best epoch is 55... figures...</p><p>** This version has some issues with the stomach, so I prompt for a pillow there. A higher rez and lower LR will probably do better for the next version.</p><hr /><p></p><h2 id="training"><span style="color:rgb(250, 82, 82)">Training</span></h2><p>Want to contribute? Check out my <a target="_blank" rel="ugc" href="https://civarchive.com/articles/13187/hunyuan-video-lora-training">Training article</a>!</p><p></p><h2 id="disclaimer"><span style="color:rgb(253, 126, 20)">Disclaimer</span></h2><p><strong>AI</strong> doesn't know any better, but we do.</p>

Front Doggystyle

front-doggy-wan22-LN-e46-az420.safetensors

Files with SHA256: 8f6c26223d7b3a8209b0d5b8e659293302e621f835dd1630679b06d7f23df839