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 |
|
|
@@ -486,13 +488,15 @@ InternVL 2.0 是一个多模态大语言模型系列，包含各种规模的模
|
|
| 486 |
|
| 487 |
### 视频相关评测
|
| 488 |
|
| 489 |
-
|
|
| 490 |
-
|
|
| 491 |
-
|
|
| 492 |
-
|
|
| 493 |
-
|
|
| 494 |
-
| Video
|
| 495 |
-
| Video
|
|
|
|
|
|
|
| 496 |
|
| 497 |
- 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。
|
| 498 |
|
|
|
|
| 68 |
|
| 69 |
### Video Benchmarks
|
| 70 |
|
| 71 |
+
| Benchmark | GPT-4V | VILA-1.5 | LLaVA-NeXT-Video | InternVL2-40B | InternVL2-Llama3-76B |
|
| 72 |
+
| :-------------------------: | :----: | :------: | :--------------: | :-----------: | :------------------: |
|
| 73 |
+
| Model Size | - | 34B | 34B | 40B | 76B |
|
| 74 |
+
| | | | | | |
|
| 75 |
+
| MVBench | - | - | - | 72.5 | 69.6 |
|
| 76 |
+
| MMBench-Video<sub>8f</sub> | 1.53 | - | - | 1.32 | 1.37 |
|
| 77 |
+
| MMBench-Video<sub>16f</sub> | 1.68 | - | - | 1.45 | 1.52 |
|
| 78 |
+
| Video-MME<br>wo subs | 59.9 | 59.0 | 52.0 | TODO | TODO |
|
| 79 |
+
| Video-MME<br>w/ subs | 63.3 | 59.4 | 54.9 | 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 |
|
|
|
|
| 488 |
|
| 489 |
### 视频相关评测
|
| 490 |
|
| 491 |
+
| 评测数据集 | GPT-4V | VILA-1.5 | LLaVA-NeXT-Video | InternVL2-40B | InternVL2-Llama3-76B |
|
| 492 |
+
| :-------------------------: | :----: | :------: | :--------------: | :-----------: | :------------------: |
|
| 493 |
+
| 模型大小 | - | 34B | 34B | 40B | 76B |
|
| 494 |
+
| | | | | | |
|
| 495 |
+
| MVBench | - | - | - | 72.5 | 69.6 |
|
| 496 |
+
| MMBench-Video<sub>8f</sub> | 1.53 | - | - | 1.32 | 1.37 |
|
| 497 |
+
| MMBench-Video<sub>16f</sub> | 1.68 | - | - | 1.45 | 1.52 |
|
| 498 |
+
| Video-MME<br>wo subs | 59.9 | 59.0 | 52.0 | TODO | TODO |
|
| 499 |
+
| Video-MME<br>w/ subs | 63.3 | 59.4 | 54.9 | TODO | TODO |
|
| 500 |
|
| 501 |
- 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。
|
| 502 |
|