| """DPGBench evaluator for T2I evaluation.""" |
|
|
| import torch |
| import numpy as np |
| from PIL import Image |
| from typing import List |
|
|
| from dpg_evaluator import MPLUG, load_prompt2id, load_dpg_metadata, evaluate_batch |
|
|
|
|
| class DPGEvaluator: |
| """Lazy-loaded DPGBench model for computing image-text alignment scores.""" |
|
|
| def __init__(self, device: str = "cuda"): |
| self.device = device |
| self.model = None |
| self.prompt2id = None |
| self.question_dict = None |
|
|
| def _ensure_loaded(self): |
| if self.model is None: |
| self.model = MPLUG(device=self.device) |
| if self.prompt2id is None: |
| self.prompt2id = load_prompt2id() |
| if self.question_dict is None: |
| self.question_dict = load_dpg_metadata() |
|
|
| @torch.no_grad() |
| def compute_batch_scores(self, images_np: np.ndarray, prompts: List[str]) -> torch.Tensor: |
| """ |
| Compute DPGBench scores for a batch of image-prompt pairs. |
| |
| Args: |
| images_np: numpy array of images with shape (B, H, W, C), values in [0, 255] |
| prompts: list of prompt strings, length B |
| |
| Returns: |
| torch.Tensor of shape (B,) containing DPGBench score per sample |
| """ |
| self._ensure_loaded() |
|
|
| |
| valid_indices = [] |
| valid_images = [] |
| valid_prompts = [] |
| for i, img in enumerate(images_np): |
| if img.shape[0] > 0 and img.shape[1] > 0: |
| valid_indices.append(i) |
| valid_images.append(Image.fromarray(img)) |
| valid_prompts.append(prompts[i]) |
| else: |
| print(f"[DPGBench] Warning: Skipping image {i} with invalid shape {img.shape}") |
|
|
| |
| if not valid_images: |
| return torch.zeros(len(images_np)) |
|
|
| results = evaluate_batch( |
| images=valid_images, |
| prompts=valid_prompts, |
| prompt2id=self.prompt2id, |
| question_dict=self.question_dict, |
| vqa_fn=self.model.batch_vqa, |
| ) |
|
|
| |
| scores = torch.zeros(len(images_np)) |
| for idx, result in zip(valid_indices, results): |
| scores[idx] = result['score'] |
| return scores |
|
|
