Upload folder using huggingface_hub
Browse files
README.md
CHANGED
|
@@ -68,13 +68,15 @@ InternVL 2.0 is a multimodal large language model series, featuring models of va
|
|
| 68 |
|
| 69 |
### Video Benchmarks
|
| 70 |
|
| 71 |
-
|
|
| 72 |
-
|
|
| 73 |
-
|
|
| 74 |
-
|
|
| 75 |
-
|
|
| 76 |
-
| Video
|
| 77 |
-
| Video
|
|
|
|
|
|
|
| 78 |
|
| 79 |
- We evaluate our models on MVBench by extracting 16 frames from each video, and each frame was resized to a 448x448 image.
|
| 80 |
|
|
@@ -606,13 +608,15 @@ InternVL 2.0 是一个多模态大语言模型系列，包含各种规模的模
|
|
| 606 |
|
| 607 |
### 视频相关评测
|
| 608 |
|
| 609 |
-
|
|
| 610 |
-
|
|
| 611 |
-
|
|
| 612 |
-
|
|
| 613 |
-
|
|
| 614 |
-
| Video
|
| 615 |
-
| Video
|
|
|
|
|
|
|
| 616 |
|
| 617 |
- 我们通过从每个视频中提取16帧来评估我们的模型在MVBench上的性能，每个视频帧被调整为448x448的图像。
|
| 618 |
|
|
|
|
| 68 |
|
| 69 |
### Video Benchmarks
|
| 70 |
|
| 71 |
+
| Benchmark | VideoChat2-Phi3 | VideoChat2-HD-Mistral | Mini-InternVL-2B-1-5 | InternVL2-2B |
|
| 72 |
+
| :-------------------------: | :-------------: | :-------------------: | :------------------: | :----------: |
|
| 73 |
+
| Model Size | 4B | 7B | 2.2B | 2.2B |
|
| 74 |
+
| | | | | |
|
| 75 |
+
| MVBench | 55.1 | 60.4 | 37.0 | 60.2 |
|
| 76 |
+
| MMBench-Video<sub>8f</sub> | - | - | 0.99 | 0.97 |
|
| 77 |
+
| MMBench-Video<sub>16f</sub> | - | - | 1.04 | 1.03 |
|
| 78 |
+
| Video-MME<br>wo subs | - | 42.3 | TODO | TODO |
|
| 79 |
+
| Video-MME<br>w/ subs | - | 54.6 | TODO | TODO |
|
| 80 |
|
| 81 |
- We evaluate our models on MVBench by extracting 16 frames from each video, and each frame was resized to a 448x448 image.
|
| 82 |
|
|
|
|
| 608 |
|
| 609 |
### 视频相关评测
|
| 610 |
|
| 611 |
+
| 评测数据集 | VideoChat2-Phi3 | VideoChat2-HD-Mistral | Mini-InternVL-2B-1-5 | InternVL2-2B |
|
| 612 |
+
| :-------------------------: | :-------------: | :-------------------: | :------------------: | :----------: |
|
| 613 |
+
| 模型大小 | 4B | 7B | 2.2B | 2.2B |
|
| 614 |
+
| | | | | |
|
| 615 |
+
| MVBench | 55.1 | 60.4 | 37.0 | 60.2 |
|
| 616 |
+
| MMBench-Video<sub>8f</sub> | - | - | 0.99 | 0.97 |
|
| 617 |
+
| MMBench-Video<sub>16f</sub> | - | - | 1.04 | 1.03 |
|
| 618 |
+
| Video-MME<br>wo subs | - | 42.3 | TODO | TODO |
|
| 619 |
+
| Video-MME<br>w/ subs | - | 54.6 | TODO | TODO |
|
| 620 |
|
| 621 |
- 我们通过从每个视频中提取16帧来评估我们的模型在MVBench上的性能，每个视频帧被调整为448x448的图像。
|
| 622 |
|