CamSFT-4B

Camera-movement understanding model, supervised fine-tuned from Qwen/Qwen3-VL-4B-Instruct. Given a video, it outputs structured JSON describing every camera-movement segment โ€” time span, basic-movement type / direction / speed, and special techniques.

Usage

Recommended โ€” the CamDistill repo

The CamDistill repo provides a one-line entry point that applies the official prompt and the exact video settings used for training and evaluation:

python camera_movement_sft/infer_single.py \
    --model ddz16/CamSFT-4B \
    --video /path/to/video.mp4

With ๐Ÿค— Transformers

CamSFT is a standard Qwen3-VL model, so it can be loaded directly:

from transformers import Qwen3VLForConditionalGeneration, AutoProcessor

model = Qwen3VLForConditionalGeneration.from_pretrained(
    "ddz16/CamSFT-4B", dtype="bfloat16", device_map="auto"
)
processor = AutoProcessor.from_pretrained("ddz16/CamSFT-4B")

The exact system/user prompt and the video preprocessing (fps, max frames, resolution) are provided in the CamDistill repo; using them is required to reproduce the paper's results.

Downloads last month
32
Safetensors
Model size
4B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for ddz16/CamSFT-4B

Finetuned
(565)
this model
Quantizations
1 model

Paper for ddz16/CamSFT-4B