SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.16969 · 2024

ZoomLDM: Latent Diffusion Model for multi-scale image generation

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 20 functions out of this paper's own repositories and ran 10 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
cvlab-stonybrook/ZoomLDM canonical 10 of 20
FunctionStatusWhere it lives
box Ran cvlab-stonybrook/ZoomLDM/large_image_gen/resizer.py
code served (permissive licence) · get_code("372b275922a9f531")
create_logger Ran cvlab-stonybrook/ZoomLDM/cdm_dit/train_cdm.py
code served (permissive licence) · get_code("b2cc78f4103df806")
cubic Ran cvlab-stonybrook/ZoomLDM/large_image_gen/resizer.py
code served (permissive licence) · get_code("ee77cd8fde21ffd1")
default Ran cvlab-stonybrook/ZoomLDM/ldm/modules/attention.py
code served (permissive licence) · get_code("424012cb37b31172")
exists Ran cvlab-stonybrook/ZoomLDM/ldm/modules/attention.py
code served (permissive licence) · get_code("aa5486a3650902d8")
get_1d_sincos_pos_embed_from_grid Ran cvlab-stonybrook/ZoomLDM/cdm_dit/models.py
code served (permissive licence) · get_code("e5947aba1d10885f")
ismap Ran cvlab-stonybrook/ZoomLDM/ldm/util.py
code served (permissive licence) · get_code("d72762b700feee6f")
lanczos2 Ran cvlab-stonybrook/ZoomLDM/large_image_gen/resizer.py
code served (permissive licence) · get_code("f106699fc610a684")
modulate Ran cvlab-stonybrook/ZoomLDM/cdm_dit/models.py
code served (permissive licence) · get_code("03310bba324ae4fb")
uniq Ran cvlab-stonybrook/ZoomLDM/ldm/modules/attention.py
code served (permissive licence) · get_code("9a299fe5ae09e407")
autocast Not yet run cvlab-stonybrook/ZoomLDM/ldm/util.py
code served (permissive licence) · get_code("32af1c12fd826ef2")
collate_fn Not yet run cvlab-stonybrook/ZoomLDM/utils.py
code served (permissive licence) · get_code("97c840d12ce5cf07")
decode_large_image Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/utils.py
code served (permissive licence) · get_code("8163165d266649ff")
gaussian_kernel Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/utils.py
code served (permissive licence) · get_code("cd927cf05827a83d")
get_conditioning Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/postprocess.py
code served (permissive licence) · get_code("ec487ba24ac66507")
log_txt_as_img Not yet run cvlab-stonybrook/ZoomLDM/ldm/util.py
code served (permissive licence) · get_code("f9bd2e83191afad1")
model_pred Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/utils.py
code served (permissive licence) · get_code("a37c1537ad066a85")
my_collate Not yet run cvlab-stonybrook/ZoomLDM/cdm_dit/train_cdm.py
code served (permissive licence) · get_code("865a41ede071b2cb")
postprocess_image Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/postprocess.py
code served (permissive licence) · get_code("34bbbbc4a2ae919a")
slerp Not yet run cvlab-stonybrook/ZoomLDM/large_image_gen/postprocess.py
code served (permissive licence) · get_code("ecc1c08d5ab92766")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Diffusion models have revolutionized image generation, yet several challenges restrict their application to large-image domains, such as digital pathology and satellite imagery. Given that it is infeasible to directly train a model on 'whole' images from domains with potential gigapixel sizes, diffusion-based generative methods have focused on synthesizing small, fixed-size patches extracted from these images. However, generating small patches has limited applicability since patch-based models fail to capture the global structures and wider context of large images, which can be crucial for synthesizing (semantically) accurate samples. To overcome this limitation, we present ZoomLDM, a diffusion model tailored for generating images across multiple scales. Central to our approach is a novel magnification-aware conditioning mechanism that utilizes self-supervised learning (SSL) embeddings and allows the diffusion model to synthesize images at different 'zoom' levels, i.e., fixed-size patches extracted from large images at varying scales. ZoomLDM synthesizes coherent histopathology images that remain contextually accurate and detailed at different zoom levels, achieving state-of-the-art image generation quality across all scales and excelling in the data-scarce setting of generating thumbnails of entire large images. The multi-scale nature of ZoomLDM unlocks additional capabilities in large image generation, enabling computationally tractable and globally coherent image synthesis up to $4096 \times 4096$ pixels and $4\times$ super-resolution. Additionally, multi-scale features extracted from ZoomLDM are highly effective in multiple instance learning experiments.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2411.16969")
get_code_for_paper("2411.16969")
have("2411.16969")

Connect an agent — have() is free.