Files
ComfyUI-Tween/gimm_vfi_arch/generalizable_INR/configs.py
Ethanfel d642255e70 Add GIMM-VFI support (NeurIPS 2024) with single-pass arbitrary-timestep interpolation
Integrates GIMM-VFI alongside existing BIM/EMA/SGM models. Key feature: generates
all intermediate frames in one forward pass (no recursive 2x passes needed for 4x/8x).

- Vendor gimm_vfi_arch/ from kijai/ComfyUI-GIMM-VFI with device fixes
- Two variants: RAFT-based (~80MB) and FlowFormer-based (~123MB)
- Auto-download checkpoints from HuggingFace (Kijai/GIMM-VFI_safetensors)
- Three new nodes: Load GIMM-VFI Model, GIMM-VFI Interpolate, GIMM-VFI Segment Interpolate
- single_pass toggle: True=arbitrary timestep (default), False=recursive like other models
- ds_factor parameter for high-res input downscaling

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-13 13:11:45 +01:00

58 lines
2.1 KiB
Python

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
# --------------------------------------------------------
# References:
# ginr-ipc: https://github.com/kakaobrain/ginr-ipc
# --------------------------------------------------------
from typing import List, Optional
from dataclasses import dataclass, field
from omegaconf import OmegaConf, MISSING
from .modules.module_config import HypoNetConfig
@dataclass
class GIMMConfig:
type: str = "gimm"
ema: Optional[bool] = None
ema_value: Optional[float] = None
fwarp_type: str = "linear"
hyponet: HypoNetConfig = field(default_factory=HypoNetConfig)
coord_range: List[float] = MISSING
modulated_layer_idxs: Optional[List[int]] = None
@classmethod
def create(cls, config):
# We need to specify the type of the default DataEncoderConfig.
# Otherwise, data_encoder will be initialized & structured as "unfold" type (which is default value)
# hence merging with the config with other type would cause config error.
defaults = OmegaConf.structured(cls(ema=False))
config = OmegaConf.merge(defaults, config)
return config
@dataclass
class GIMMVFIConfig:
type: str = "gimmvfi"
ema: Optional[bool] = None
ema_value: Optional[float] = None
fwarp_type: str = "linear"
rec_weight: float = 0.1
hyponet: HypoNetConfig = field(default_factory=HypoNetConfig)
raft_iter: int = 20
coord_range: List[float] = MISSING
modulated_layer_idxs: Optional[List[int]] = None
@classmethod
def create(cls, config):
# We need to specify the type of the default DataEncoderConfig.
# Otherwise, data_encoder will be initialized & structured as "unfold" type (which is default value)
# hence merging with the config with other type would cause config error.
defaults = OmegaConf.structured(cls(ema=False))
config = OmegaConf.merge(defaults, config)
return config