SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2109.02974 · ICCV · 2021

FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting

Hongsheng Li, Lewei Lu, Xiaogang Wang, Jifeng Dai, Rui Liu, Xiaoyu Shi, Hanming Deng, Wenxiu Sun, Yangyi Huang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 11 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
ruiliu-ai/FuseFormer — 10 of 11
FunctionStatusWhere it lives
AddPosEmb Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("974fc45de281481c")
Attention Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("a702c7ce4cea9802")
BaseNetwork Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("6a1c3869290d0d57")
Encoder Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("5231e4794aac485a")
FusionFeedForward Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("3d843c10ed59c635")
MultiHeadedAttention Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("634deecea407ff83")
SoftComp Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("b99e0f5e4538ded3")
SoftSplit Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("2e02b61d792f9198")
TransformerBlock Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("afa615aa2f60ae16")
deconv Ran ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("e73d218a85e7f736")
InpaintGenerator Not yet run ruiliu-ai/FuseFormer/model/fuseformer.py
pointer only (licence: NONE) · get_code("dbafafc1baa0fc1c")

Repositories linked to this paper

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

Abstract

Transformer, as a strong and flexible architecture for modelling long-range relations, has been widely explored in vision tasks. However, when used in video inpainting that requires fine-grained representation, existed method

For agents

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

get_harvested_code_for_paper("2109.02974")
get_code_for_paper("2109.02974")
have("2109.02974")

Connect an agent — have() is free.