Rename package directory from fmpose to fmpose3d

README.md

@@ -28,9 +28,9 @@ conda activate fmpose_3d
 ```bash
 git clone xxxx.git  # clone this repo
 # TestPyPI (pre-release/testing build)
-pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ fmpose==0.0.5
+pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ fmpose3d==0.0.5
 # Future Official PyPI release
-# pip install fmpose
+# pip install fmpose3d
 ```
 
 ## Demo 

animals/demo/vis_animals.py

@@ -13,8 +13,8 @@
 from PIL import Image
 import matplotlib.gridspec as gridspec
 import imageio
-from fmpose.animals.common.arguments import opts as parse_args
-from fmpose.common.camera import normalize_screen_coordinates, camera_to_world
+from fmpose3d.animals.common.arguments import opts as parse_args
+from fmpose3d.common.camera import normalize_screen_coordinates, camera_to_world
 
 sys.path.append(os.getcwd())
 
@@ -37,7 +37,7 @@
     CFM = getattr(module, "Model")
 else:
     # Load model from installed fmpose package
-    from fmpose.models import Model as CFM
+    from fmpose3d.models import Model as CFM
 
 from deeplabcut.pose_estimation_pytorch.apis import superanimal_analyze_images
 

animals/models/model_animals.py

@@ -6,7 +6,7 @@
 from einops import rearrange
 from timm.models.layers import DropPath
 
-from fmpose.animals.models.graph_frames import Graph
+from fmpose3d.animals.models.graph_frames import Graph
 
 
 class TimeEmbedding(nn.Module):

animals/scripts/main_animal3d.py

@@ -6,9 +6,9 @@
 import numpy as np
 from tqdm import tqdm
 import torch.optim as optim
-from fmpose.animals.common.arguments import opts as parse_args
-from fmpose.animals.common.utils import *
-from fmpose.animals.common.animal3d_dataset import TrainDataset
+from fmpose3d.animals.common.arguments import opts as parse_args
+from fmpose3d.animals.common.utils import *
+from fmpose3d.animals.common.animal3d_dataset import TrainDataset
 import time
 
 args = parse_args().parse()
@@ -30,7 +30,7 @@
     CFM = getattr(module, "Model")
 else:
     # Load model from installed fmpose package
-    from fmpose.animals.models import Model as CFM
+    from fmpose3d.animals.models import Model as CFM
 
 def train(opt, actions, train_loader, model, optimizer, epoch):
     return step('train', opt, actions, train_loader, model, optimizer, epoch)

demo/vis_in_the_wild.py

@@ -10,12 +10,12 @@
 sys.path.append(os.getcwd())
 
 # Auto-download checkpoint files if missing
-from fmpose.lib.checkpoint.download_checkpoints import ensure_checkpoints
+from fmpose3d.lib.checkpoint.download_checkpoints import ensure_checkpoints
 ensure_checkpoints()
 
-from fmpose.lib.preprocess import h36m_coco_format, revise_kpts
-from fmpose.lib.hrnet.gen_kpts import gen_video_kpts as hrnet_pose
-from fmpose.common.arguments import opts as parse_args
+from fmpose3d.lib.preprocess import h36m_coco_format, revise_kpts
+from fmpose3d.lib.hrnet.gen_kpts import gen_video_kpts as hrnet_pose
+from fmpose3d.common.arguments import opts as parse_args
 
 args = parse_args().parse()
 os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu
@@ -30,7 +30,7 @@
     spec.loader.exec_module(module)
     CFM = getattr(module, 'Model')
 
-from fmpose.common.camera import *
+from fmpose3d.common.camera import *
 
 import matplotlib
 import matplotlib.pyplot as plt 

fmpose/aggregation_methods.py → fmpose3d/aggregation_methods.py

@@ -1,5 +1,5 @@
 import torch
-from fmpose.common.utils import project_to_2d
+from fmpose3d.common.utils import project_to_2d
 
 def average_aggregation(list_hypothesis):
     return torch.mean(torch.stack(list_hypothesis), dim=0)

fmpose/animals/models/model_animal3d.py → fmpose3d/animals/models/model_animal3d.py

@@ -6,7 +6,7 @@
 from einops import rearrange
 from timm.models.layers import DropPath
 
-from fmpose.animals.models.graph_frames import Graph
+from fmpose3d.animals.models.graph_frames import Graph
 
 class TimeEmbedding(nn.Module):
     def __init__(self, dim: int, hidden_dim: int = 64):

fmpose/common/h36m_dataset.py → fmpose3d/common/h36m_dataset.py

@@ -2,9 +2,9 @@
 
 import numpy as np
 
-from fmpose.common.camera import normalize_screen_coordinates
-from fmpose.common.mocap_dataset import MocapDataset
-from fmpose.common.skeleton import Skeleton
+from fmpose3d.common.camera import normalize_screen_coordinates
+from fmpose3d.common.mocap_dataset import MocapDataset
+from fmpose3d.common.skeleton import Skeleton
 
 h36m_skeleton = Skeleton(
     parents=[

fmpose/common/load_data_hm36.py → fmpose3d/common/load_data_hm36.py

@@ -1,9 +1,9 @@
 import numpy as np
 import torch.utils.data as data
 
-from fmpose.common.camera import normalize_screen_coordinates, world_to_camera
-from fmpose.common.generator import ChunkedGenerator
-from fmpose.common.utils import deterministic_random
+from fmpose3d.common.camera import normalize_screen_coordinates, world_to_camera
+from fmpose3d.common.generator import ChunkedGenerator
+from fmpose3d.common.utils import deterministic_random
 
 
 class Fusion(data.Dataset):

fmpose/lib/hrnet/gen_kpts.py → fmpose3d/lib/hrnet/gen_kpts.py

@@ -15,22 +15,22 @@
 import cv2
 import copy
 
-from fmpose.lib.hrnet.lib.utils.utilitys import plot_keypoint, PreProcess, write, load_json
-from fmpose.lib.hrnet.lib.config import cfg, update_config
-from fmpose.lib.hrnet.lib.utils.transforms import *
-from fmpose.lib.hrnet.lib.utils.inference import get_final_preds
-from fmpose.lib.hrnet.lib.models import pose_hrnet
+from fmpose3d.lib.hrnet.lib.utils.utilitys import plot_keypoint, PreProcess, write, load_json
+from fmpose3d.lib.hrnet.lib.config import cfg, update_config
+from fmpose3d.lib.hrnet.lib.utils.transforms import *
+from fmpose3d.lib.hrnet.lib.utils.inference import get_final_preds
+from fmpose3d.lib.hrnet.lib.models import pose_hrnet
 
 cfg_dir = osp.join(osp.dirname(osp.abspath(__file__)), 'experiments') + '/'
 
 # Auto-download checkpoints if missing and get checkpoint paths
-from fmpose.lib.checkpoint.download_checkpoints import ensure_checkpoints, get_checkpoint_path
+from fmpose3d.lib.checkpoint.download_checkpoints import ensure_checkpoints, get_checkpoint_path
 ensure_checkpoints()
 
 # Loading human detector model
-from fmpose.lib.yolov3.human_detector import load_model as yolo_model
-from fmpose.lib.yolov3.human_detector import yolo_human_det as yolo_det
-from fmpose.lib.sort.sort import Sort
+from fmpose3d.lib.yolov3.human_detector import load_model as yolo_model
+from fmpose3d.lib.yolov3.human_detector import yolo_human_det as yolo_det
+from fmpose3d.lib.sort.sort import Sort
 
 def parse_args():
     parser = argparse.ArgumentParser(description='Train keypoints network')

fmpose/lib/hrnet/lib/utils/utilitys.py → fmpose3d/lib/hrnet/lib/utils/utilitys.py

@@ -3,9 +3,9 @@
 import torch
 import json
 import torchvision.transforms as transforms
-from fmpose.lib.hrnet.lib.utils.transforms import *
+from fmpose3d.lib.hrnet.lib.utils.transforms import *
 
-from fmpose.lib.hrnet.lib.utils.coco_h36m import coco_h36m
+from fmpose3d.lib.hrnet.lib.utils.coco_h36m import coco_h36m
 import numpy as np
 
 joint_pairs = [[0, 1], [1, 3], [0, 2], [2, 4],

fmpose/lib/yolov3/darknet.py → fmpose3d/lib/yolov3/darknet.py

@@ -8,8 +8,8 @@
 import os
 import sys
 
-from fmpose.lib.yolov3.util import convert2cpu as cpu
-from fmpose.lib.yolov3.util import predict_transform
+from fmpose3d.lib.yolov3.util import convert2cpu as cpu
+from fmpose3d.lib.yolov3.util import predict_transform
 
 
 class test_net(nn.Module):

fmpose/lib/yolov3/human_detector.py → fmpose3d/lib/yolov3/human_detector.py

@@ -9,10 +9,10 @@
 import pickle as pkl
 import argparse
 
-from fmpose.lib.yolov3.util import *
-from fmpose.lib.yolov3.darknet import Darknet
-from fmpose.lib.yolov3 import preprocess
-from fmpose.lib.checkpoint.download_checkpoints import get_checkpoint_path
+from fmpose3d.lib.yolov3.util import *
+from fmpose3d.lib.yolov3.darknet import Darknet
+from fmpose3d.lib.yolov3 import preprocess
+from fmpose3d.lib.checkpoint.download_checkpoints import get_checkpoint_path
 
 cur_dir = os.path.dirname(os.path.realpath(__file__))
 project_root = os.path.join(cur_dir, '../../../')

fmpose/lib/yolov3/util.py → fmpose3d/lib/yolov3/util.py

@@ -4,8 +4,8 @@
 import numpy as np
 import cv2
 import os.path as osp
-from fmpose.lib.yolov3.bbox import bbox_iou
-from fmpose.lib.checkpoint.download_checkpoints import get_checkpoint_dir
+from fmpose3d.lib.yolov3.bbox import bbox_iou
+from fmpose3d.lib.checkpoint.download_checkpoints import get_checkpoint_dir
 
 
 def get_path(cur_file):

fmpose/models/model_GAMLP.py → fmpose3d/models/model_GAMLP.py

@@ -4,7 +4,7 @@
 import torch.nn as nn
 import math
 from einops import rearrange
-from fmpose.models.graph_frames import Graph
+from fmpose3d.models.graph_frames import Graph
 from functools import partial
 from einops import rearrange, repeat
 from timm.models.layers import DropPath

pyproject.toml

@@ -54,7 +54,7 @@ include = ["fmpose*"]
 "*" = ["*.yaml", "*.cfg", "*.names"]
 
 [tool.setuptools.dynamic]
-version = {attr = "fmpose.__version__"}
+version = {attr = "fmpose3d.__version__"}
 
 [tool.black]
 line-length = 100

scripts/FMPose3D_main.py

@@ -9,10 +9,10 @@
 import torch.optim as optim
 from tqdm import tqdm
 
-from fmpose.common import opts, Human36mDataset, Fusion
-from fmpose.common.utils import *
+from fmpose3d.common import opts, Human36mDataset, Fusion
+from fmpose3d.common.utils import *
 
-from fmpose.aggregation_methods import aggregation_RPEA_weighted_by_2D_error
+from fmpose3d.aggregation_methods import aggregation_RPEA_weighted_by_2D_error
 
 args = opts().parse()
 os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu
@@ -33,7 +33,7 @@
     CFM = getattr(module, "Model")
 else:
     # Load model from installed fmpose package
-    from fmpose.models import Model as CFM
+    from fmpose3d.models import Model as CFM
 
 
 def test_multi_hypothesis(

tests/test_demo_human.py

@@ -25,8 +25,8 @@ def test_output_dir(tmp_path):
 def test_2d_pose_estimation(test_image_path, test_output_dir):
     """Test that 2D pose estimation runs and produces output."""
     # Import here to avoid import issues at collection time
-    from fmpose.lib.hrnet.gen_kpts import gen_video_kpts as hrnet_pose
-    from fmpose.lib.preprocess import h36m_coco_format, revise_kpts
+    from fmpose3d.lib.hrnet.gen_kpts import gen_video_kpts as hrnet_pose
+    from fmpose3d.lib.preprocess import h36m_coco_format, revise_kpts
 
     # Run 2D pose estimation
     keypoints, scores = hrnet_pose(test_image_path, det_dim=416, num_peroson=1, gen_output=True, type='image')
@@ -52,9 +52,9 @@ def test_2d_pose_estimation(test_image_path, test_output_dir):
 def test_demo_pipeline_runs(test_image_path):
     """Test that the full demo pipeline can be imported and key components work."""
     # Test imports
-    from fmpose.lib.hrnet.gen_kpts import gen_video_kpts
-    from fmpose.lib.preprocess import h36m_coco_format, revise_kpts
-    from fmpose.models import Model
+    from fmpose3d.lib.hrnet.gen_kpts import gen_video_kpts
+    from fmpose3d.lib.preprocess import h36m_coco_format, revise_kpts
+    from fmpose3d.models import Model
 
     assert gen_video_kpts is not None
     assert h36m_coco_format is not None

tests/test_model.py

@@ -1,6 +1,6 @@
 import pytest
 import torch
-from fmpose.models import Model
+from fmpose3d.models import Model
 
 class Args:
     """Mock args for model configuration."""

tests/test_training_pipeline.py

@@ -1,7 +1,7 @@
 import pytest
 import torch
 import torch.optim as optim
-from fmpose.models import Model
+from fmpose3d.models import Model
 
 
 class Args: