[KongchangAI]
Model

ProtoDiffusion

A diffusion model for multi-layer graphic design generation that compresses representations into HW+N tokens via shared spatial prototypes and per-layer candidate vectors, supports image- and text-conditioned generation, and can infer layer layouts.

Timeline (last 90 days)

Nov 20

ProtoDiffusion proposes a multi-layer image generation representation: all layers share a single spatial prototype, and each layer retains only one candidate vector, thereby compressing the representation of an N-layer design to HW+N tokens (before backbone-specific packing), whereas a full-grid scheme requires NHW tokens.

Unverified50%
Nov 20

ProtoDiffusion performs joint denoising in a compact latent space, simultaneously denoising the shared prototype and a fixed number of candidate slots, with conditions being either images or text and bounding boxes serving as optional layout guidance.

Unverified50%
Nov 20

ProtoDiffusion's candidate slots can be left empty, and the model decides on its own which slots are empty, so the number of layers need not be provided as input.

Unverified50%
Nov 20

ProtoDiffusion can also infer layer layout when no bounding boxes are provided, with bounding boxes acting only as optional constraints rather than required inputs.

Unverified50%
Nov 20

The same ProtoDiffusion architecture supports both image-to-design and text-to-design tasks, without needing separate networks designed for the two directions.

Unverified50%
Nov 20

On the LayerD evaluation set, ProtoDiffusion's image-to-design RGB L1 score is 0.040 with bounding boxes and 0.057 without bounding boxes.

Unverified50%
Nov 20

On the LayerD evaluation set, ProtoDiffusion achieves an alpha soft IoU of 0.850 under box guidance, comparable to the strongest fine-tuned baseline of 0.871.

Unverified50%
Nov 20

In the text-to-design direction on LayerD, ProtoDiffusion achieves an FID of 66.41 on merged images, compared with 86.84 for FLUX.2 [klein].

Unverified50%
Nov 20

ProtoDiffusion achieves a text accuracy score of 0.860 in the text-to-design direction.

Unverified50%
Nov 20

ProtoDiffusion is 19.1 times faster than the same-backbone 'generate-then-decompose' cascade scheme, with the speed gap mainly stemming from eliminating the layer decomposition step.

Unverified50%

3 more timeline events

All Facts (13)

Unverified

ProtoDiffusion proposes a multi-layer image generation representation: all layers share a single spatial prototype, and each layer retains only one candidate vector, thereby compressing the representation of an N-layer design to HW+N tokens (before backbone-specific packing), whereas a full-grid scheme requires NHW tokens.

50%
Unverified

ProtoDiffusion performs joint denoising in a compact latent space, simultaneously denoising the shared prototype and a fixed number of candidate slots, with conditions being either images or text and bounding boxes serving as optional layout guidance.

50%
Unverified

ProtoDiffusion's candidate slots can be left empty, and the model decides on its own which slots are empty, so the number of layers need not be provided as input.

50%
Unverified

ProtoDiffusion can also infer layer layout when no bounding boxes are provided, with bounding boxes acting only as optional constraints rather than required inputs.

50%
Unverified

The same ProtoDiffusion architecture supports both image-to-design and text-to-design tasks, without needing separate networks designed for the two directions.

50%
Unverified

On the LayerD evaluation set, ProtoDiffusion's image-to-design RGB L1 score is 0.040 with bounding boxes and 0.057 without bounding boxes.

50%
Unverified

On the LayerD evaluation set, ProtoDiffusion achieves an alpha soft IoU of 0.850 under box guidance, comparable to the strongest fine-tuned baseline of 0.871.

50%
Unverified

In the text-to-design direction on LayerD, ProtoDiffusion achieves an FID of 66.41 on merged images, compared with 86.84 for FLUX.2 [klein].

50%
Unverified

ProtoDiffusion achieves a text accuracy score of 0.860 in the text-to-design direction.

50%
Unverified

ProtoDiffusion is 19.1 times faster than the same-backbone 'generate-then-decompose' cascade scheme, with the speed gap mainly stemming from eliminating the layer decomposition step.

50%
Unverified

Existing cascade-style multi-layer generation schemes (generate the whole image first, then decompose layers) accumulate errors across two stages, whereas ProtoDiffusion jointly models all layers within a single model, with spatial alignment between layers naturally constrained by the shared prototype.

50%
Unverified

ProtoDiffusion's shared-prototype assumption posits that all layers share the same set of spatial structure priors; for designs with greatly differing spatial distributions across layers, whether this sharing is sufficient still requires validation in more scenarios.

50%
Unverified

A fixed number of candidate slots means that designs with an extremely large number of layers must rely on slot reuse or expansion strategies.

50%

Source Articles