Does not work correctly by API

[slonce70]

Good afternoon, the problem is the following - in the browser, using the module, changes clothes perfectly. Unniffed the request, prescribed parameters in the api http://127.0.0.1:7860/sdapi/v1/img2img - does not change clothes fully, it feels like denosing does not set, although set 0.9. Here are the parameters that I send on the api

{
    "prompt": "<lora:add_detail:0.8>, <lora:more_details:0.3>, masterpiece, high quality, 8k, beautiful lighting, absurdres, perfect anatomy, ((swimsuit, blue swimsuit)), ",
    "negative_prompt": "(worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark",
    "sampler_name": "Euler a",
    "styles": [],
    "seed": -1,
    "subseed": -1,
    "subseed_strength": 0,
    "seed_resize_from_h": -1,
    "seed_resize_from_w": -1,
    "batch_size": 1,
    "n_iter": 1,
    "steps": 24,
    "cfg_scale": 7,
    "width": 640,
    "height": 848,
    "restore_faces": false,
    "tiling": false,
    "do_not_save_samples": false,
    "do_not_save_grid": false,
    "eta": 0,
    "denoising_strength": 0.9,
    "s_min_uncond": 0,
    "s_churn": 0,
    "s_tmax": 0,
    "s_tmin": 0,
    "s_noise": 0,
    "override_settings": {},
    "disable_extra_networks": false,
    "comments": {},
    "init_images": [
        "base64 our images"
    ],
    "send_images": true,
    "save_images": false,
    "alwayson_scripts": {
        "A Person Mask Generator": {
            "args": [
                true,
                [
                    "clothes"
                ],
                true,
                4,
                0,
                1,
                false, (or true - not change nothing)
                256,
                8
            ]
        }
    }
}

Been racking my brain for a couple days now that I haven't changed it already. What is the problem and what to do, please tell me. Thanks in advance

The first image is taken via the web. Models etc. are the same, all parameters are the same

This is the result we get from the API

[meyiapir]

It's the same problem, only mine is blurry

[djbielejeski]

Everything is working fine for me over the API

Web UI

Request

{
    "prompt": "masterpiece, high quality, 8k, beautiful lighting, absurdres, ((white wedding dress))",
    "negative_prompt": "((deformed)), (cross-eyed), (closed eyes), (bad anatomy), disfigured, (mutation), (mutated), (extra limbs), 3d render, cartoon",
    "sampler_name": "Euler a",
    "styles": [],
    "seed": 1645829012,
    "subseed": -1,
    "subseed_strength": 0,
    "seed_resize_from_h": -1,
    "seed_resize_from_w": -1,
    "batch_size": 1,
    "n_iter": 1,
    "steps": 20,
    "cfg_scale": 7,
    "width": 448,
    "height": 576,
    "restore_faces": false,
    "tiling": false,
    "do_not_save_samples": false,
    "do_not_save_grid": false,
    "eta": 0,
    "denoising_strength": 0.75,
    "s_min_uncond": 0,
    "s_churn": 0,
    "s_tmax": 0,
    "s_tmin": 0,
    "s_noise": 0,
    "override_settings": {},
    "disable_extra_networks": false,
    "comments": {},
    "init_images": [
         "data:image/jpeg;base64,/9j/4AAQSkZJR...snipped..."
    ],
    "send_images": true,
    "save_images": false,
    "alwayson_scripts": {
        "A Person Mask Generator": {
            "args": [
                true,
                [
                    "clothes"
                ],
                true,
                4,
                0,
                1,
                0,
                256,
                8,
                8
            ]
        }
    }
}

API Response

https://base64.guru/converter/decode/image

{
    "images": [
        "iVBOR...snipped..."
    ],
    "parameters": {
        "prompt": "masterpiece, high quality, 8k, beautiful lighting, absurdres, ((white wedding dress))",
        "negative_prompt": "((deformed)), (cross-eyed), (closed eyes), (bad anatomy), disfigured, (mutation), (mutated), (extra limbs), 3d render, cartoon",
        "styles": [],
        "seed": 1645829012,
        "subseed": -1,
        "subseed_strength": 0,
        "seed_resize_from_h": -1,
        "seed_resize_from_w": -1,
        "sampler_name": "Euler a",
        "batch_size": 1,
        "n_iter": 1,
        "steps": 20,
        "cfg_scale": 7,
        "width": 448,
        "height": 576,
        "restore_faces": false,
        "tiling": false,
        "do_not_save_samples": false,
        "do_not_save_grid": false,
        "eta": 0,
        "denoising_strength": 0.75,
        "s_min_uncond": 0,
        "s_churn": 0,
        "s_tmax": 0,
        "s_tmin": 0,
        "s_noise": 0,
        "override_settings": {},
        "override_settings_restore_afterwards": true,
        "refiner_checkpoint": null,
        "refiner_switch_at": null,
        "disable_extra_networks": false,
        "comments": {},
        "init_images": null,
        "resize_mode": 0,
        "image_cfg_scale": null,
        "mask": null,
        "mask_blur_x": 4,
        "mask_blur_y": 4,
        "mask_blur": null,
        "inpainting_fill": 0,
        "inpaint_full_res": true,
        "inpaint_full_res_padding": 0,
        "inpainting_mask_invert": 0,
        "initial_noise_multiplier": null,
        "latent_mask": null,
        "sampler_index": "Euler",
        "include_init_images": false,
        "script_name": null,
        "script_args": [],
        "send_images": true,
        "save_images": false,
        "alwayson_scripts": {
            "A Person Mask Generator": {
                "args": [
                    true,
                    [
                        "clothes"
                    ],
                    true,
                    4,
                    0,
                    1,
                    0,
                    256,
                    8,
                    8
                ]
            }
        }
    },
    "info": "{\"prompt\": \"masterpiece, high quality, 8k, beautiful lighting, absurdres, ((white wedding dress))\", \"all_prompts\": [\"masterpiece, high quality, 8k, beautiful lighting, absurdres, ((white wedding dress))\"], \"negative_prompt\": \"((deformed)), (cross-eyed), (closed eyes), (bad anatomy), disfigured, (mutation), (mutated), (extra limbs), 3d render, cartoon\", \"all_negative_prompts\": [\"((deformed)), (cross-eyed), (closed eyes), (bad anatomy), disfigured, (mutation), (mutated), (extra limbs), 3d render, cartoon\"], \"seed\": 1645829012, \"all_seeds\": [1645829012], \"subseed\": 200682659, \"all_subseeds\": [200682659], \"subseed_strength\": 0.0, \"width\": 448, \"height\": 576, \"sampler_name\": \"Euler a\", \"cfg_scale\": 7.0, \"steps\": 20, \"batch_size\": 1, \"restore_faces\": false, \"face_restoration_model\": null, \"sd_model_name\": \"v1-5-pruned-emaonly-pruned\", \"sd_model_hash\": \"ad5749cdfa\", \"sd_vae_name\": \"vae-ft-mse-840000-ema-pruned.vae.pt\", \"sd_vae_hash\": \"ff724e9611\", \"seed_resize_from_w\": -1, \"seed_resize_from_h\": -1, \"denoising_strength\": 0.75, \"extra_generation_params\": {\"Eta\": 0.0}, \"index_of_first_image\": 0, \"infotexts\": [\"masterpiece, high quality, 8k, beautiful lighting, absurdres, ((white wedding dress))\nNegative prompt: ((deformed)), (cross-eyed), (closed eyes), (bad anatomy), disfigured, (mutation), (mutated), (extra limbs), 3d render, cartoon\nSteps: 20, Sampler: Euler a, CFG scale: 7.0, Seed: 1645829012, Size: 448x576, Model hash: ad5749cdfa, Model: v1-5-pruned-emaonly-pruned, VAE hash: ff724e9611, VAE: vae-ft-mse-840000-ema-pruned.vae.pt, Denoising strength: 0.75, Eta: 0.0, Version: v1.7.0\"], \"styles\": [], \"job_timestamp\": \"20240106111833\", \"clip_skip\": 1, \"is_using_inpainting_conditioning\": false, \"version\": \"v1.7.0\"}"
}

[djbielejeski]

Here is the image I used

[slonce70]

Checked, the problem remains with complex photos. Here I am sending a full request in your likeness

{
    "prompt": "((swimsuit, blue swimsuit))",
    "negative_prompt": "(worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark",
    "sampler_name": "Euler a",
    "styles": [],
    "seed": 1645829012,
    "subseed": -1,
    "subseed_strength": 0,
    "seed_resize_from_h": -1,
    "seed_resize_from_w": -1,
    "batch_size": 1,
    "n_iter": 1,
    "steps": 20,
    "cfg_scale": 7,
    "width": 640,
    "height": 848,
    "restore_faces": false,
    "tiling": false,
    "do_not_save_samples": false,
    "do_not_save_grid": false,
    "eta": 0,
    "denoising_strength": 0.9,
    "s_min_uncond": 0,
    "s_churn": 0,
    "s_tmax": 0,
    "s_tmin": 0,
    "s_noise": 0,
    "override_settings": {},
    "disable_extra_networks": false,
    "comments": {},
    "init_images": [
        "data:image/jpeg;base64,"
    ],
    "send_images": true,
    "save_images": false,
    "alwayson_scripts": {
        "A Person Mask Generator": {
            "args": [
                true,
                [
                    "clothes"
                ],
                true,
                4,
                0,
                1,
                0,
                256,
                8,
                8
            ]
        }
    }
}

Its original images

This image is processed through the browser with the same parameters - see the result - fully functional without problems

And this is the image generated through your API request. As you can see, it seems that APL draws the mask incorrectly for some reason.

As you can see it didn't process all sections, which in the browser is done without problems. The problem is not only with this photo, but with many of them.

Here is the response code I'm getting

{"images":["],
    "parameters": {
        "prompt": "((swimsuit, blue swimsuit))",
        "negative_prompt": "(worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark",
        "styles": [],
        "seed": 1645829012,
        "subseed": -1,
        "subseed_strength": 0.0,
        "seed_resize_from_h": -1,
        "seed_resize_from_w": -1,
        "sampler_name": "Euler a",
        "batch_size": 1,
        "n_iter": 1,
        "steps": 20,
        "cfg_scale": 7.0,
        "width": 640,
        "height": 848,
        "restore_faces": false,
        "tiling": false,
        "do_not_save_samples": false,
        "do_not_save_grid": false,
        "eta": 0.0,
        "denoising_strength": 0.9,
        "s_min_uncond": 0.0,
        "s_churn": 0.0,
        "s_tmax": 0.0,
        "s_tmin": 0.0,
        "s_noise": 0.0,
        "override_settings": {},
        "override_settings_restore_afterwards": true,
        "refiner_checkpoint": null,
        "refiner_switch_at": null,
        "disable_extra_networks": false,
        "comments": {},
        "init_images": null,
        "resize_mode": 0,
        "image_cfg_scale": null,
        "mask": null,
        "mask_blur_x": 4,
        "mask_blur_y": 4,
        "mask_blur": null,
        "inpainting_fill": 0,
        "inpaint_full_res": true,
        "inpaint_full_res_padding": 0,
        "inpainting_mask_invert": 0,
        "initial_noise_multiplier": null,
        "latent_mask": null,
        "sampler_index": "Euler",
        "include_init_images": false,
        "script_name": null,
        "script_args": [],
        "send_images": true,
        "save_images": false,
        "alwayson_scripts": {
            "A Person Mask Generator": {
                "args": [
                    true,
                    [
                        "clothes"
                    ],
                    true,
                    4,
                    0,
                    1,
                    0,
                    256,
                    8,
                    8
                ]
            }
        }
    },
    "info": "{\"prompt\": \"((swimsuit, blue swimsuit))\", \"all_prompts\": [\"((swimsuit, blue swimsuit))\"], \"negative_prompt\": \"(worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark\", \"all_negative_prompts\": [\"(worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark\"], \"seed\": 1645829012, \"all_seeds\": [1645829012], \"subseed\": 900395898, \"all_subseeds\": [900395898], \"subseed_strength\": 0.0, \"width\": 640, \"height\": 848, \"sampler_name\": \"Euler a\", \"cfg_scale\": 7.0, \"steps\": 20, \"batch_size\": 1, \"restore_faces\": false, \"face_restoration_model\": null, \"sd_model_name\": \"epicrealism_pureEvolutionV5-inpainting\", \"sd_model_hash\": \"caa95002a1\", \"sd_vae_name\": \"vae-ft-mse-840000-ema-pruned.ckpt\", \"sd_vae_hash\": \"c6a580b13a\", \"seed_resize_from_w\": -1, \"seed_resize_from_h\": -1, \"denoising_strength\": 0.9, \"extra_generation_params\": {\"Eta\": 0.0}, \"index_of_first_image\": 0, \"infotexts\": [\"((swimsuit, blue swimsuit))\\nNegative prompt: (worst quality, low quality, ugly:1.4), (deformed, distorted, disfigured:1.3), stacked torsos, totem pole, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, text, watermark\\nSteps: 20, Sampler: Euler a, CFG scale: 7.0, Seed: 1645829012, Size: 640x848, Model hash: caa95002a1, Model: epicrealism_pureEvolutionV5-inpainting, VAE hash: c6a580b13a, VAE: vae-ft-mse-840000-ema-pruned.ckpt, Denoising strength: 0.9, Conditional mask weight: 1.0, Eta: 0.0, Version: v1.7.0\"], \"styles\": [], \"job_timestamp\": \"20240106233951\", \"clip_skip\": 1, \"is_using_inpainting_conditioning\": true, \"version\": \"v1.7.0\"}"
}

[slonce70]

Here's another example.

Original photo

Made through the browser - ((white dress))

Made via API - with the same parameters and promt ((white dress)).

[MkKsko]

@djbielejeski Hello! Is there a way to extract the mask via API? The resulting json only contains the generated image and the mask is nowhere to be found.

[monstari]

@djbielejeski Hello! Is there a way to extract the mask via API? The resulting json only contains the generated image and the mask is nowhere to be found.

Same! That would be usefull! I want to use this extension as negative mask.

[bog-kotikov]

Problem in color model, webui convert image to RGBA model, via api image default uses RGB. Just convert file before sending it to api server.

zh = BytesIO()
ma = Image.open('img.png').convert('RGBA')
ma.save(zh, "PNG")
img = base64.b64encode(zh.getvalue()).decode('utf-8')

[MotherSoraka]

Thank you @bog-kotikov !