dgenerate-ultralytics-headless 8.3.214__tar.gz → 8.3.216__tar.gz

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dgenerate-ultralytics-headless
-Version: 8.3.214
+Version: 8.3.216
 Summary: Automatically built Ultralytics package with python-opencv-headless dependency instead of python-opencv
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/dgenerate_ultralytics_headless.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dgenerate-ultralytics-headless
-Version: 8.3.214
+Version: 8.3.216
 Summary: Automatically built Ultralytics package with python-opencv-headless dependency instead of python-opencv
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.214"
+__version__ = "8.3.216"
 
 import importlib
 import os

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/cfg/datasets/coco-pose.yaml

@@ -22,6 +22,27 @@ flip_idx: [0, 2, 1, 4, 3, 6, 5, 8, 7, 10, 9, 12, 11, 14, 13, 16, 15]
 names:
   0: person
 
+# Keypoint names per class
+kpt_names:
+  0:
+    - nose
+    - left_eye
+    - right_eye
+    - left_ear
+    - right_ear
+    - left_shoulder
+    - right_shoulder
+    - left_elbow
+    - right_elbow
+    - left_wrist
+    - right_wrist
+    - left_hip
+    - right_hip
+    - left_knee
+    - right_knee
+    - left_ankle
+    - right_ankle
+
 # Download script/URL (optional)
 download: |
   from pathlib import Path

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/cfg/datasets/coco8-pose.yaml

@@ -22,5 +22,26 @@ flip_idx: [0, 2, 1, 4, 3, 6, 5, 8, 7, 10, 9, 12, 11, 14, 13, 16, 15]
 names:
   0: person
 
+# Keypoint names per class
+kpt_names:
+  0:
+    - nose
+    - left_eye
+    - right_eye
+    - left_ear
+    - right_ear
+    - left_shoulder
+    - right_shoulder
+    - left_elbow
+    - right_elbow
+    - left_wrist
+    - right_wrist
+    - left_hip
+    - right_hip
+    - left_knee
+    - right_knee
+    - left_ankle
+    - right_ankle
+
 # Download script/URL (optional)
 download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco8-pose.zip

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/cfg/datasets/dog-pose.yaml

@@ -20,5 +20,33 @@ kpt_shape: [24, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y
 names:
   0: dog
 
+# Keypoint names per class
+kpt_names:
+  0:
+    - front_left_paw
+    - front_left_knee
+    - front_left_elbow
+    - rear_left_paw
+    - rear_left_knee
+    - rear_left_elbow
+    - front_right_paw
+    - front_right_knee
+    - front_right_elbow
+    - rear_right_paw
+    - rear_right_knee
+    - rear_right_elbow
+    - tail_start
+    - tail_end
+    - left_ear_base
+    - right_ear_base
+    - nose
+    - chin
+    - left_ear_tip
+    - right_ear_tip
+    - left_eye
+    - right_eye
+    - withers
+    - throat
+
 # Download script/URL (optional)
 download: https://github.com/ultralytics/assets/releases/download/v0.0.0/dog-pose.zip

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/cfg/datasets/hand-keypoints.yaml

@@ -22,5 +22,30 @@ flip_idx:
 names:
   0: hand
 
+# Keypoint names per class
+kpt_names:
+  0:
+    - wrist
+    - thumb_cmc
+    - thumb_mcp
+    - thumb_ip
+    - thumb_tip
+    - index_mcp
+    - index_pip
+    - index_dip
+    - index_tip
+    - middle_mcp
+    - middle_pip
+    - middle_dip
+    - middle_tip
+    - ring_mcp
+    - ring_pip
+    - ring_dip
+    - ring_tip
+    - pinky_mcp
+    - pinky_pip
+    - pinky_dip
+    - pinky_tip
+
 # Download script/URL (optional)
 download: https://github.com/ultralytics/assets/releases/download/v0.0.0/hand-keypoints.zip

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/cfg/datasets/tiger-pose.yaml

@@ -21,5 +21,21 @@ flip_idx: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
 names:
   0: tiger
 
+# Keypoint names per class
+kpt_names:
+  0:
+    - nose
+    - head
+    - withers
+    - tail_base
+    - right_hind_hock
+    - right_hind_paw
+    - left_hind_paw
+    - left_hind_hock
+    - right_front_wrist
+    - right_front_paw
+    - left_front_wrist
+    - left_front_paw
+
 # Download script/URL (optional)
 download: https://github.com/ultralytics/assets/releases/download/v0.0.0/tiger-pose.zip

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/engine/exporter.py

@@ -385,7 +385,7 @@ class Exporter:
             assert not tflite or not ARM64 or not LINUX, "TFLite export with NMS unsupported on ARM64 Linux"
             assert not is_tf_format or TORCH_1_13, "TensorFlow exports with NMS require torch>=1.13"
             assert not onnx or TORCH_1_13, "ONNX export with NMS requires torch>=1.13"
-            if getattr(model, "end2end", False):
+            if getattr(model, "end2end", False) or isinstance(model.model[-1], RTDETRDecoder):
                 LOGGER.warning("'nms=True' is not available for end2end models. Forcing 'nms=False'.")
                 self.args.nms = False
             self.args.conf = self.args.conf or 0.25  # set conf default value for nms export
@@ -502,6 +502,8 @@ class Exporter:
             self.metadata["dla"] = dla  # make sure `AutoBackend` uses correct dla device if it has one
         if model.task == "pose":
             self.metadata["kpt_shape"] = model.model[-1].kpt_shape
+            if hasattr(model, "kpt_names"):
+                self.metadata["kpt_names"] = model.kpt_names
 
         LOGGER.info(
             f"\n{colorstr('PyTorch:')} starting from '{file}' with input shape {tuple(im.shape)} BCHW and "
@@ -1039,7 +1041,7 @@ class Exporter:
             attempt_download_asset(f"{onnx2tf_file}.zip", unzip=True, delete=True)
 
         # Export to ONNX
-        if "rtdetr" in self.model.model[-1]._get_name().lower():
+        if isinstance(self.model.model[-1], RTDETRDecoder):
             self.args.opset = self.args.opset or 19
             assert 16 <= self.args.opset <= 19, "RTDETR export requires opset>=16;<=19"
         self.args.simplify = True

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/models/sam/build.py

@@ -11,6 +11,7 @@ from functools import partial
 import torch
 
 from ultralytics.utils.downloads import attempt_download_asset
+from ultralytics.utils.torch_utils import TORCH_1_13
 
 from .modules.decoders import MaskDecoder
 from .modules.encoders import FpnNeck, Hiera, ImageEncoder, ImageEncoderViT, MemoryEncoder, PromptEncoder
@@ -207,7 +208,7 @@ def _build_sam(
     if checkpoint is not None:
         checkpoint = attempt_download_asset(checkpoint)
         with open(checkpoint, "rb") as f:
-            state_dict = torch.load(f)
+            state_dict = torch.load(f, weights_only=False) if TORCH_1_13 else torch.load(f)
         sam.load_state_dict(state_dict)
     sam.eval()
     return sam
@@ -302,7 +303,7 @@ def _build_sam2(
     if checkpoint is not None:
         checkpoint = attempt_download_asset(checkpoint)
         with open(checkpoint, "rb") as f:
-            state_dict = torch.load(f)["model"]
+            state_dict = (torch.load(f, weights_only=False) if TORCH_1_13 else torch.load(f))["model"]
         sam2.load_state_dict(state_dict)
     sam2.eval()
     return sam2

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/models/yolo/pose/train.py

@@ -91,6 +91,11 @@ class PoseTrainer(yolo.detect.DetectionTrainer):
         """Set keypoints shape attribute of PoseModel."""
         super().set_model_attributes()
         self.model.kpt_shape = self.data["kpt_shape"]
+        kpt_names = self.data.get("kpt_names")
+        if not kpt_names:
+            names = list(map(str, range(self.model.kpt_shape[0])))
+            kpt_names = {i: names for i in range(self.model.nc)}
+        self.model.kpt_names = kpt_names
 
     def get_validator(self):
         """Return an instance of the PoseValidator class for validation."""

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/models/yolo/segment/predict.py

@@ -108,6 +108,7 @@ class SegmentationPredictor(DetectionPredictor):
             masks = ops.process_mask(proto, pred[:, 6:], pred[:, :4], img.shape[2:], upsample=True)  # HWC
             pred[:, :4] = ops.scale_boxes(img.shape[2:], pred[:, :4], orig_img.shape)
         if masks is not None:
-            keep = masks.sum((-2, -1)) > 0  # only keep predictions with masks
-            pred, masks = pred[keep], masks[keep]
+            keep = masks.amax((-2, -1)) > 0  # only keep predictions with masks
+            if not all(keep):  # most predictions have masks
+                pred, masks = pred[keep], masks[keep]  # indexing is slow
         return Results(orig_img, path=img_path, names=self.model.names, boxes=pred[:, :6], masks=masks)

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/nn/autobackend.py

@@ -585,7 +585,7 @@ class AutoBackend(nn.Module):
             for k, v in metadata.items():
                 if k in {"stride", "batch", "channels"}:
                     metadata[k] = int(v)
-                elif k in {"imgsz", "names", "kpt_shape", "args"} and isinstance(v, str):
+                elif k in {"imgsz", "names", "kpt_shape", "kpt_names", "args"} and isinstance(v, str):
                     metadata[k] = eval(v)
             stride = metadata["stride"]
             task = metadata["task"]
@@ -593,6 +593,7 @@ class AutoBackend(nn.Module):
             imgsz = metadata["imgsz"]
             names = metadata["names"]
             kpt_shape = metadata.get("kpt_shape")
+            kpt_names = metadata.get("kpt_names")
             end2end = metadata.get("args", {}).get("nms", False)
             dynamic = metadata.get("args", {}).get("dynamic", dynamic)
             ch = metadata.get("channels", 3)

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/utils/export/__init__.py

@@ -8,6 +8,7 @@ from pathlib import Path
 import torch
 
 from ultralytics.utils import IS_JETSON, LOGGER
+from ultralytics.utils.torch_utils import TORCH_2_4
 
 from .imx import torch2imx  # noqa
 
@@ -36,6 +37,7 @@ def torch2onnx(
     Notes:
         Setting `do_constant_folding=True` may cause issues with DNN inference for torch>=1.12.
     """
+    kwargs = {"dynamo": False} if TORCH_2_4 else {}
     torch.onnx.export(
         torch_model,
         im,
@@ -46,6 +48,7 @@ def torch2onnx(
         input_names=input_names,
         output_names=output_names,
         dynamic_axes=dynamic or None,
+        **kwargs,
     )
 
 

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/utils/ops.py

@@ -517,12 +517,19 @@ def crop_mask(masks, boxes):
     Returns:
         (torch.Tensor): Cropped masks.
     """
-    _, h, w = masks.shape
-    x1, y1, x2, y2 = torch.chunk(boxes[:, :, None], 4, 1)  # x1 shape(n,1,1)
-    r = torch.arange(w, device=masks.device, dtype=x1.dtype)[None, None, :]  # rows shape(1,1,w)
-    c = torch.arange(h, device=masks.device, dtype=x1.dtype)[None, :, None]  # cols shape(1,h,1)
-
-    return masks * ((r >= x1) * (r < x2) * (c >= y1) * (c < y2))
+    n, h, w = masks.shape
+    if n < 50:  # faster for fewer masks (predict)
+        for i, (x1, y1, x2, y2) in enumerate(boxes.round().int()):
+            masks[i, :y1] = 0
+            masks[i, y2:] = 0
+            masks[i, :, :x1] = 0
+            masks[i, :, x2:] = 0
+        return masks
+    else:  # faster for more masks (val)
+        x1, y1, x2, y2 = torch.chunk(boxes[:, :, None], 4, 1)  # x1 shape(n,1,1)
+        r = torch.arange(w, device=masks.device, dtype=x1.dtype)[None, None, :]  # rows shape(1,1,w)
+        c = torch.arange(h, device=masks.device, dtype=x1.dtype)[None, :, None]  # cols shape(1,h,1)
+        return masks * ((r >= x1) * (r < x2) * (c >= y1) * (c < y2))
 
 
 def process_mask(protos, masks_in, bboxes, shape, upsample: bool = False):
@@ -541,20 +548,15 @@ def process_mask(protos, masks_in, bboxes, shape, upsample: bool = False):
             are the height and width of the input image. The mask is applied to the bounding boxes.
     """
     c, mh, mw = protos.shape  # CHW
-    ih, iw = shape
     masks = (masks_in @ protos.float().view(c, -1)).view(-1, mh, mw)  # CHW
-    width_ratio = mw / iw
-    height_ratio = mh / ih
 
-    downsampled_bboxes = bboxes.clone()
-    downsampled_bboxes[:, 0] *= width_ratio
-    downsampled_bboxes[:, 2] *= width_ratio
-    downsampled_bboxes[:, 3] *= height_ratio
-    downsampled_bboxes[:, 1] *= height_ratio
+    width_ratio = mw / shape[1]
+    height_ratio = mh / shape[0]
+    ratios = torch.tensor([[width_ratio, height_ratio, width_ratio, height_ratio]], device=bboxes.device)
 
-    masks = crop_mask(masks, downsampled_bboxes)  # CHW
+    masks = crop_mask(masks, boxes=bboxes * ratios)  # CHW
     if upsample:
-        masks = F.interpolate(masks[None], shape, mode="bilinear", align_corners=False)[0]  # CHW
+        masks = F.interpolate(masks[None], shape, mode="bilinear")[0]  # CHW
     return masks.gt_(0.0)
 
 
@@ -600,7 +602,7 @@ def scale_masks(masks, shape, padding: bool = True):
     top, left = (int(round(pad_h - 0.1)), int(round(pad_w - 0.1))) if padding else (0, 0)
     bottom = mh - int(round(pad_h + 0.1))
     right = mw - int(round(pad_w + 0.1))
-    return F.interpolate(masks[..., top:bottom, left:right], shape, mode="bilinear", align_corners=False)  # NCHW masks
+    return F.interpolate(masks[..., top:bottom, left:right], shape, mode="bilinear")  # NCHW masks
 
 
 def scale_coords(img1_shape, coords, img0_shape, ratio_pad=None, normalize: bool = False, padding: bool = True):

{dgenerate_ultralytics_headless-8.3.214 → dgenerate_ultralytics_headless-8.3.216}/ultralytics/utils/plotting.py

@@ -384,25 +384,32 @@ class Annotator:
                 overlay[mask.astype(bool)] = colors[i]
             self.im = cv2.addWeighted(self.im, 1 - alpha, overlay, alpha, 0)
         else:
-            assert isinstance(masks, torch.Tensor), "`masks` must be a torch.Tensor if `im_gpu` is provided."
+            assert isinstance(masks, torch.Tensor), "'masks' must be a torch.Tensor if 'im_gpu' is provided."
             if len(masks) == 0:
                 self.im[:] = im_gpu.permute(1, 2, 0).contiguous().cpu().numpy() * 255
+                return
             if im_gpu.device != masks.device:
                 im_gpu = im_gpu.to(masks.device)
+
+            ih, iw = self.im.shape[:2]
+            if not retina_masks:
+                # Use scale_masks to properly remove padding and upsample, convert bool to float first
+                masks = ops.scale_masks(masks[None].float(), (ih, iw))[0] > 0.5
+                # Convert original BGR image to RGB tensor
+                im_gpu = (
+                    torch.from_numpy(self.im).to(masks.device).permute(2, 0, 1).flip(0).contiguous().float() / 255.0
+                )
+
             colors = torch.tensor(colors, device=masks.device, dtype=torch.float32) / 255.0  # shape(n,3)
             colors = colors[:, None, None]  # shape(n,1,1,3)
             masks = masks.unsqueeze(3)  # shape(n,h,w,1)
             masks_color = masks * (colors * alpha)  # shape(n,h,w,3)
-
             inv_alpha_masks = (1 - masks * alpha).cumprod(0)  # shape(n,h,w,1)
             mcs = masks_color.max(dim=0).values  # shape(n,h,w,3)
 
-            im_gpu = im_gpu.flip(dims=[0])  # flip channel
-            im_gpu = im_gpu.permute(1, 2, 0).contiguous()  # shape(h,w,3)
+            im_gpu = im_gpu.flip(dims=[0]).permute(1, 2, 0).contiguous()  # shape(h,w,3)
             im_gpu = im_gpu * inv_alpha_masks[-1] + mcs
-            im_mask = im_gpu * 255
-            im_mask_np = im_mask.byte().cpu().numpy()
-            self.im[:] = im_mask_np if retina_masks else ops.scale_image(im_mask_np, self.im.shape)
+            self.im[:] = (im_gpu * 255).byte().cpu().numpy()
         if self.pil:
             # Convert im back to PIL and update draw
             self.fromarray(self.im)