Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,50 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: image-text-to-text
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
<br>
|
| 7 |
+
<br>
|
| 8 |
+
|
| 9 |
+
# Spatial-LLaVA-7B Model Card
|
| 10 |
+
|
| 11 |
+
## 🤖 Model details
|
| 12 |
+
|
| 13 |
+
**Model type:**
|
| 14 |
+
|
| 15 |
+
This finetuned LLaVA model is trained from [liuhaotian/llava-pretrain-vicuna-7b-v1.3](https://huggingface.co/liuhaotian/llava-pretrain-vicuna-7b-v1.3) for improving spatial relation reasoning of large multi-modal model.
|
| 16 |
+
|
| 17 |
+
LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data.
|
| 18 |
+
It is an auto-regressive language model, based on the transformer architecture.
|
| 19 |
+
|
| 20 |
+
## 🎯 Intended use
|
| 21 |
+
**Primary intended uses:**
|
| 22 |
+
The primary use of LLaVA is research on large multimodal models and chatbots.
|
| 23 |
+
|
| 24 |
+
**Primary intended users:**
|
| 25 |
+
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence.
|
| 26 |
+
|
| 27 |
+
## 📚 Training dataset
|
| 28 |
+
Instruction following training: [rogerxi/LLaVA-Spatial-Instruct-850K](https://huggingface.co/datasets/rogerxi/LLaVA-Spatial-Instruct-850K)
|
| 29 |
+
|
| 30 |
+
## 📊 Evaluation
|
| 31 |
+
A collection of 10 benchmarks:
|
| 32 |
+
| Model | VQAv2 | GQA | VizWiz | SQA | TextVQA | POPE | MME | MM-Bench | MM-Bench-cn | MM-Vet |
|
| 33 |
+
|:----------------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:----------:|:--------:|:-----------:|:--------:|
|
| 34 |
+
| LLaVA-1.5-7b | 78.5 | 62.0 | **50.0** | 66.8 | 58.2 | 85.9 | **1510.7** | 64.3 | 58.3 | 31.1 |
|
| 35 |
+
| Spatial-LLaVA-7b | **79.7** | **62.7** | 48.7 | **68.7** | **58.5** | **87.2** | 1472.7 | **67.8** | **60.7** | **31.6** |
|
| 36 |
+
|
| 37 |
+
[SpatialRGPT-Bench](https://huggingface.co/datasets/a8cheng/SpatialRGPT-Bench) (with placeholder replaced by object name):
|
| 38 |
+
### Qualitative Spatial Relations
|
| 39 |
+
|
| 40 |
+
| Model | Below/Above | Left/Right | Big/Small | Tall/Short | Wide/Thin | Behind/Front | Avg |
|
| 41 |
+
|:-----------------------:|:------------:|:-----------:|:----------:|:-----------:|:----------:|:-------------:|:-------------: |
|
| 42 |
+
| LLaVA-1.5-7b | 53.91 | 53.49 | 45.36 | 40.00 | **50.00** | 51.04 | 48.97 |
|
| 43 |
+
| Spatial-LLaVA-7b | **56.32** | **66.28** | **60.82** | **48.57** | 49.02 | **52.08** | **55.12** |
|
| 44 |
+
|
| 45 |
+
### Quantitative Spatial Relations
|
| 46 |
+
|
| 47 |
+
| Model | Direct Dist (m / ratio) | Horizontal Dist (m / ratio) | Vertical Dist (m / ratio) | Width (m / ratio) | Height (m / ratio) | Direction (° / ratio) |
|
| 48 |
+
|:-----------------------:|:------------------------:|:----------------------------:|:--------------------------:|:------------------:|:-------------------:|:----------------------:|
|
| 49 |
+
| LLaVA-1.5-7b | 12.90 / 1.06 | 10.68 / 2.03 | 20.79 / 0.94 | **24.19 / 0.50** | 14.29 / 5.27 | 10.23 / 58.33 |
|
| 50 |
+
| Spatial-LLaVA-7b | **24.19 / 0.57** | **14.56 / 0.62** | **41.58 / 0.42** | 22.58 / 1.12 | **18.25 / 2.92** | **20.45 / 56.47** |
|