update pipeline files

2026-02-07 13:50:38 -08:00
parent 59e75f81d3
commit 0c41d1dab5
4 changed files with 102 additions and 127 deletions
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: Apache-2.0
+"""HunyuanGameCraft pipeline implementations."""
@@ -0,0 +1,100 @@
+# SPDX-License-Identifier: Apache-2.0
+"""
+HunyuanGameCraft video diffusion pipeline implementation.
+
+This module implements the HunyuanGameCraft pipeline for camera/action-conditioned
+video generation with the modular pipeline architecture.
+"""
+
+from fastvideo.fastvideo_args import FastVideoArgs
+from fastvideo.logger import init_logger
+from fastvideo.pipelines.composed_pipeline_base import ComposedPipelineBase
+from fastvideo.pipelines.stages import (
+    ConditioningStage,
+    DecodingStage,
+    InputValidationStage,
+    LatentPreparationStage,
+    TextEncodingStage,
+    TimestepPreparationStage,
+)
+from fastvideo.pipelines.stages.gamecraft_denoising import GameCraftDenoisingStage
+
+logger = init_logger(__name__)
+
+
+class HunyuanGameCraftPipeline(ComposedPipelineBase):
+    """
+    Pipeline for HunyuanGameCraft video generation.
+    
+    This pipeline supports:
+    - Text-to-video generation with camera/action conditioning
+    - Autoregressive generation with history frames
+    - 33-channel input (16 latent + 16 gt_latent + 1 mask)
+    - CameraNet for encoding Plücker coordinates
+    """
+
+    _required_config_modules = [
+        "text_encoder",
+        "text_encoder_2",
+        "tokenizer",
+        "tokenizer_2",
+        "vae",
+        "transformer",
+        "scheduler",
+    ]
+
+    def create_pipeline_stages(self, fastvideo_args: FastVideoArgs):
+        """Set up pipeline stages with proper dependency injection."""
+
+        self.add_stage(
+            stage_name="input_validation_stage",
+            stage=InputValidationStage(),
+        )
+
+        self.add_stage(
+            stage_name="prompt_encoding_stage_primary",
+            stage=TextEncodingStage(
+                text_encoders=[
+                    self.get_module("text_encoder"),
+                    self.get_module("text_encoder_2"),
+                ],
+                tokenizers=[
+                    self.get_module("tokenizer"),
+                    self.get_module("tokenizer_2"),
+                ],
+            ),
+        )
+
+        self.add_stage(
+            stage_name="conditioning_stage",
+            stage=ConditioningStage(),
+        )
+
+        self.add_stage(
+            stage_name="timestep_preparation_stage",
+            stage=TimestepPreparationStage(scheduler=self.get_module("scheduler")),
+        )
+
+        self.add_stage(
+            stage_name="latent_preparation_stage",
+            stage=LatentPreparationStage(
+                scheduler=self.get_module("scheduler"),
+                transformer=self.get_module("transformer"),
+            ),
+        )
+
+        self.add_stage(
+            stage_name="denoising_stage",
+            stage=GameCraftDenoisingStage(
+                transformer=self.get_module("transformer"),
+                scheduler=self.get_module("scheduler"),
+            ),
+        )
+
+        self.add_stage(
+            stage_name="decoding_stage",
+            stage=DecodingStage(vae=self.get_module("vae")),
+        )
+
+
+EntryClass = HunyuanGameCraftPipeline
@@ -1,8 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""HunyuanGameCraft pipeline module."""
-
-from fastvideo.pipelines.basic.hunyuangamecraft.hunyuangamecraft_pipeline import (
-    HunyuanGameCraftPipeline,
-)
-
-__all__ = ["HunyuanGameCraftPipeline"]
@@ -1,119 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""
-HunyuanGameCraft video diffusion pipeline implementation.
-
-This module contains an implementation of the HunyuanGameCraft video diffusion pipeline
-using the modular pipeline architecture. HunyuanGameCraft extends HunyuanVideo with
-camera pose conditioning via CameraNet using Plücker coordinate representation.
-"""
-
-from fastvideo.fastvideo_args import FastVideoArgs
-from fastvideo.logger import init_logger
-from fastvideo.models.schedulers.scheduling_flow_match_euler_discrete import (
-    FlowMatchEulerDiscreteScheduler)
-from fastvideo.pipelines.composed_pipeline_base import ComposedPipelineBase
-from fastvideo.pipelines.stages import (ConditioningStage, DecodingStage,
-                                        InputValidationStage,
-                                        LatentPreparationStage,
-                                        TextEncodingStage,
-                                        TimestepPreparationStage)
-from fastvideo.pipelines.stages.gamecraft_denoising import GameCraftDenoisingStage
-
-logger = init_logger(__name__)
-
-
-class HunyuanGameCraftPipeline(ComposedPipelineBase):
-    """
-    HunyuanGameCraft video diffusion pipeline.
-    
-    This pipeline extends the HunyuanVideo architecture with camera pose conditioning
-    for controllable game video generation. Camera poses are represented as Plücker
-    coordinates (6D ray representation) and processed through CameraNet.
-    
-    The pipeline structure is similar to HunyuanVideo but the transformer includes
-    CameraNet for camera conditioning. Camera states can be passed through the
-    ForwardBatch.camera_states field.
-    
-    Required modules:
-        - text_encoder: LLaMA-based text encoder (primary)
-        - text_encoder_2: CLIP text encoder (secondary)
-        - tokenizer: LLaMA tokenizer
-        - tokenizer_2: CLIP tokenizer
-        - vae: HunyuanVideo VAE (causal 3D)
-        - transformer: HunyuanGameCraftTransformer3DModel with CameraNet
-        - scheduler: Flow matching scheduler
-    """
-
-    _required_config_modules = [
-        "text_encoder", "text_encoder_2", "tokenizer", "tokenizer_2", "vae",
-        "transformer", "scheduler"
-    ]
-
-    def initialize_pipeline(self, fastvideo_args: FastVideoArgs):
-        """Initialize the scheduler with correct shift parameter.
-        
-        Official HunyuanGameCraft uses flow_shift=5.0 with the
-        FlowMatchDiscreteScheduler that applies the time-shift ONCE in
-        set_timesteps. The FlowMatchEulerDiscreteScheduler applies the shift
-        in __init__ (affecting sigma_min/sigma_max) AND again in set_timesteps,
-        causing a double-shift bug. To avoid this we initialise with shift=1.0
-        (identity) and then set the real shift value afterward, so only the
-        set_timesteps call applies the shift.
-        """
-        flow_shift = fastvideo_args.pipeline_config.flow_shift
-        logger.info(
-            f"Initializing FlowMatchEulerDiscreteScheduler for GameCraft "
-            f"(shift={flow_shift}, avoiding double-shift)"
-        )
-        # Create scheduler with shift=1.0 so __init__ does NOT shift
-        # sigma_min/sigma_max. Then set the real shift so set_timesteps
-        # applies it exactly once -- matching the official
-        # FlowMatchDiscreteScheduler behaviour.
-        scheduler = FlowMatchEulerDiscreteScheduler(
-            shift=1.0,
-            num_train_timesteps=1000,
-        )
-        scheduler.set_shift(flow_shift)
-        self.modules["scheduler"] = scheduler
-
-    def create_pipeline_stages(self, fastvideo_args: FastVideoArgs):
-        """Set up pipeline stages with proper dependency injection."""
-
-        self.add_stage(stage_name="input_validation_stage",
-                       stage=InputValidationStage())
-
-        self.add_stage(stage_name="prompt_encoding_stage_primary",
-                       stage=TextEncodingStage(
-                           text_encoders=[
-                               self.get_module("text_encoder"),
-                               self.get_module("text_encoder_2")
-                           ],
-                           tokenizers=[
-                               self.get_module("tokenizer"),
-                               self.get_module("tokenizer_2")
-                           ],
-                       ))
-
-        self.add_stage(stage_name="conditioning_stage",
-                       stage=ConditioningStage())
-
-        self.add_stage(stage_name="timestep_preparation_stage",
-                       stage=TimestepPreparationStage(
-                           scheduler=self.get_module("scheduler")))
-
-        self.add_stage(stage_name="latent_preparation_stage",
-                       stage=LatentPreparationStage(
-                           scheduler=self.get_module("scheduler"),
-                           transformer=self.get_module("transformer")))
-
-        self.add_stage(stage_name="denoising_stage",
-                       stage=GameCraftDenoisingStage(
-                           transformer=self.get_module("transformer"),
-                           scheduler=self.get_module("scheduler"),
-                           vae=self.get_module("vae")))
-
-        self.add_stage(stage_name="decoding_stage",
-                       stage=DecodingStage(vae=self.get_module("vae")))
-
-
-EntryClass = HunyuanGameCraftPipeline