Closed yqxuxuxu closed 5 months ago
作者您好,我使用以下代码对渲染实时性进行了测试,但是结果只有4.fps,远没有论文列出的25fps这么高,不知道是我的操作有问题还是因为卡不一样呢?
fps = [] for idx in tqdm(range(total_frames)): item = self.dataset.get_test_item(idx, source_id=view_select) data = self.fetch_data(item) data = get_novel_calib(data, self.cfg.dataset, ratio=ratio, intr_key='intr_ori', extr_key='extr_ori') start_time = time.time() with torch.no_grad(): data, _, _ = self.model(data, is_train=False) data = pts2render(data, bg_color=self.cfg.dataset.bg_color) end_time = time.time() synthesis_time = end_time - start_time fps.append(1/synthesis_time) render_novel = self.tensor2np(data['novel_view']['img_pred']) cv2.imwrite(test_path + '/%s_novel.jpg' % (data['name']), render_novel) print(sum(fps)/ len(fps))
python版本的差不多在5fps左右,实时系统需要把网络部分全部转成TensorRT,跟FloRen中类似。
好的,非常感谢。
作者您好,我使用以下代码对渲染实时性进行了测试,但是结果只有4.fps,远没有论文列出的25fps这么高,不知道是我的操作有问题还是因为卡不一样呢?