Spaces:
Running
Running
File size: 3,596 Bytes
1999a98 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
import torch
from ultralytics.data.augment import LetterBox
from ultralytics.engine.predictor import BasePredictor
from ultralytics.engine.results import Results
from ultralytics.utils import ops
class RTDETRPredictor(BasePredictor):
"""
RT-DETR (Real-Time Detection Transformer) Predictor extending the BasePredictor class for making predictions using
Baidu's RT-DETR model.
This class leverages the power of Vision Transformers to provide real-time object detection while maintaining
high accuracy. It supports key features like efficient hybrid encoding and IoU-aware query selection.
Example:
```python
from ultralytics.utils import ASSETS
from ultralytics.models.rtdetr import RTDETRPredictor
args = dict(model="rtdetr-l.pt", source=ASSETS)
predictor = RTDETRPredictor(overrides=args)
predictor.predict_cli()
```
Attributes:
imgsz (int): Image size for inference (must be square and scale-filled).
args (dict): Argument overrides for the predictor.
"""
def postprocess(self, preds, img, orig_imgs):
"""
Postprocess the raw predictions from the model to generate bounding boxes and confidence scores.
The method filters detections based on confidence and class if specified in `self.args`.
Args:
preds (list): List of [predictions, extra] from the model.
img (torch.Tensor): Processed input images.
orig_imgs (list or torch.Tensor): Original, unprocessed images.
Returns:
(list[Results]): A list of Results objects containing the post-processed bounding boxes, confidence scores,
and class labels.
"""
if not isinstance(preds, (list, tuple)): # list for PyTorch inference but list[0] Tensor for export inference
preds = [preds, None]
nd = preds[0].shape[-1]
bboxes, scores = preds[0].split((4, nd - 4), dim=-1)
if not isinstance(orig_imgs, list): # input images are a torch.Tensor, not a list
orig_imgs = ops.convert_torch2numpy_batch(orig_imgs)
results = []
for bbox, score, orig_img, img_path in zip(bboxes, scores, orig_imgs, self.batch[0]): # (300, 4)
bbox = ops.xywh2xyxy(bbox)
max_score, cls = score.max(-1, keepdim=True) # (300, 1)
idx = max_score.squeeze(-1) > self.args.conf # (300, )
if self.args.classes is not None:
idx = (cls == torch.tensor(self.args.classes, device=cls.device)).any(1) & idx
pred = torch.cat([bbox, max_score, cls], dim=-1)[idx] # filter
oh, ow = orig_img.shape[:2]
pred[..., [0, 2]] *= ow
pred[..., [1, 3]] *= oh
results.append(Results(orig_img, path=img_path, names=self.model.names, boxes=pred))
return results
def pre_transform(self, im):
"""
Pre-transforms the input images before feeding them into the model for inference. The input images are
letterboxed to ensure a square aspect ratio and scale-filled. The size must be square(640) and scaleFilled.
Args:
im (list[np.ndarray] |torch.Tensor): Input images of shape (N,3,h,w) for tensor, [(h,w,3) x N] for list.
Returns:
(list): List of pre-transformed images ready for model inference.
"""
letterbox = LetterBox(self.imgsz, auto=False, scaleFill=True)
return [letterbox(image=x) for x in im]
|