SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.16460 · ICCV · 2023

Improving Lens Flare Removal with General-Purpose Pipeline and Multiple Light Sources Recovery

Songcan Chen, Chongyi Li, Shuo Yang, Sheng-Jun Huang, Dong Liang, Yuyan Zhou

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 12 functions out of this paper's own repositories and ran 9 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
YuyanZhou1/Improving-Lens-Flare-Removal canonical 9 of 12
FunctionStatusWhere it lives
apply_blur Ran YuyanZhou1/Improving-Lens-Flare-Removal/utils.py
pointer only (licence: NONE) · get_code("b302cb7f9a0bac7d")
center_crop Ran YuyanZhou1/Improving-Lens-Flare-Removal/remove_flare.py
pointer only (licence: NONE) · get_code("195f94de9d19fd91")
get_model Ran YuyanZhou1/Improving-Lens-Flare-Removal/u_net.py
pointer only (licence: NONE) · get_code("212615acc69e2ee5")
get_scene_dataset Ran YuyanZhou1/Improving-Lens-Flare-Removal/data_provider.py
pointer only (licence: NONE) · get_code("8414ecb64760579d")
image_dataset_from_files Ran YuyanZhou1/Improving-Lens-Flare-Removal/data_provider.py
pointer only (licence: NONE) · get_code("b5ad2ce24e0594f0")
quantize_8 Ran YuyanZhou1/Improving-Lens-Flare-Removal/utils.py
pointer only (licence: NONE) · get_code("0c6930b4b2a6ec69")
remove_flare Ran YuyanZhou1/Improving-Lens-Flare-Removal/utils.py
pointer only (licence: NONE) · get_code("53635106ca0a9910")
window_partition Ran YuyanZhou1/Improving-Lens-Flare-Removal/models.py
pointer only (licence: NONE) · get_code("e2adf1ceac443f14")
window_reverse Ran YuyanZhou1/Improving-Lens-Flare-Removal/models.py
pointer only (licence: NONE) · get_code("1081e4c5bd78e72b")
build_can Not yet run YuyanZhou1/Improving-Lens-Flare-Removal/vgg.py
pointer only (licence: NONE) · get_code("45b39e353cf2a5ab")
image_dataset_from_tfrecords Not yet run YuyanZhou1/Improving-Lens-Flare-Removal/data_provider.py
pointer only (licence: NONE) · get_code("c718c80b0b30c36b")
mixup Not yet run YuyanZhou1/Improving-Lens-Flare-Removal/synthesis.py
pointer only (licence: NONE) · get_code("6cc61f5df7939b4b")

Repositories linked to this paper

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

Abstract

https://github.com/YuyanZhou1/Improving-Lens-Flare-Removal iPad Air4 Huawei Mate20 iPhone 13 pro iPhone 11 OPPO reno 4 pro iPad 2020 Input Output Figure 1. The proposed solution yields favorable results on the flare-corrupted images captured by various devices. The real and diverse flare-corrupted images are provided in our consumer electronics test dataset. The results are produced by the deep model U-Former trained using our solution that includes the new data synthesis pipeline and multiple light sources recovery strategy.

For agents

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

get_harvested_code_for_paper("2308.16460")
get_code_for_paper("2308.16460")
have("2308.16460")

Connect an agent — have() is free.