T^(h (ddlmZmZmZmZddlmZmZmZm Z m Z ddl m Z m Z erddlmZddlmZmZerddlZdd lmZdd lmZe j0eZee d Gd de Zy))AnyDictListUnion)add_end_docstringsis_torch_availableis_vision_availableloggingrequires_backends) ChunkPipelinebuild_pipeline_init_args)Image) load_image valid_imagesN)BaseModelOutput)2MODEL_FOR_ZERO_SHOT_OBJECT_DETECTION_MAPPING_NAMEST)has_image_processorc eZdZdZfdZ ddeedeeee ffdeeeefffd Z dZ ddZ d Z dd Zd d d eeeffdZxZS)ZeroShotObjectDetectionPipelinea Zero shot object detection pipeline using `OwlViTForObjectDetection`. This pipeline predicts bounding boxes of objects when you provide an image and a set of `candidate_labels`. Example: ```python >>> from transformers import pipeline >>> detector = pipeline(model="google/owlvit-base-patch32", task="zero-shot-object-detection") >>> detector( ... "http://images.cocodataset.org/val2017/000000039769.jpg", ... candidate_labels=["cat", "couch"], ... ) [{'score': 0.287, 'label': 'cat', 'box': {'xmin': 324, 'ymin': 20, 'xmax': 640, 'ymax': 373}}, {'score': 0.254, 'label': 'cat', 'box': {'xmin': 1, 'ymin': 55, 'xmax': 315, 'ymax': 472}}, {'score': 0.121, 'label': 'couch', 'box': {'xmin': 4, 'ymin': 0, 'xmax': 642, 'ymax': 476}}] >>> detector( ... "https://huggingface.co/datasets/Narsil/image_dummy/raw/main/parrots.png", ... candidate_labels=["head", "bird"], ... ) [{'score': 0.119, 'label': 'bird', 'box': {'xmin': 71, 'ymin': 170, 'xmax': 410, 'ymax': 508}}] ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial) This object detection pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"zero-shot-object-detection"`. See the list of available models on [huggingface.co/models](https://huggingface.co/models?filter=zero-shot-object-detection). c t|di||jdk(rtd|jdt |d|j ty)NtfzThe z is only available in PyTorch.vision)super__init__ framework ValueError __class__r check_model_typer)selfkwargsr s o/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/pipelines/zero_shot_object_detection.pyrz(ZeroShotObjectDetectionPipeline.__init__8sR "6" >>T !tDNN#33QRS S$) PQimagez Image.Imagecandidate_labelsc :d|vr|jd}t|ttjfr||d}nNt|tt fr5t |r*t t| dt||Dfi|S |}t| |fi|}|S)a| Detect objects (bounding boxes & classes) in the image(s) passed as inputs. Args: image (`str`, `PIL.Image` or `List[Dict[str, Any]]`): The pipeline handles three types of images: - A string containing an http url pointing to an image - A string containing a local path to an image - An image loaded in PIL directly You can use this parameter to send directly a list of images, or a dataset or a generator like so: ```python >>> from transformers import pipeline >>> detector = pipeline(model="google/owlvit-base-patch32", task="zero-shot-object-detection") >>> detector( ... [ ... { ... "image": "http://images.cocodataset.org/val2017/000000039769.jpg", ... "candidate_labels": ["cat", "couch"], ... }, ... { ... "image": "http://images.cocodataset.org/val2017/000000039769.jpg", ... "candidate_labels": ["cat", "couch"], ... }, ... ] ... ) [[{'score': 0.287, 'label': 'cat', 'box': {'xmin': 324, 'ymin': 20, 'xmax': 640, 'ymax': 373}}, {'score': 0.25, 'label': 'cat', 'box': {'xmin': 1, 'ymin': 55, 'xmax': 315, 'ymax': 472}}, {'score': 0.121, 'label': 'couch', 'box': {'xmin': 4, 'ymin': 0, 'xmax': 642, 'ymax': 476}}], [{'score': 0.287, 'label': 'cat', 'box': {'xmin': 324, 'ymin': 20, 'xmax': 640, 'ymax': 373}}, {'score': 0.254, 'label': 'cat', 'box': {'xmin': 1, 'ymin': 55, 'xmax': 315, 'ymax': 472}}, {'score': 0.121, 'label': 'couch', 'box': {'xmin': 4, 'ymin': 0, 'xmax': 642, 'ymax': 476}}]] ``` candidate_labels (`str` or `List[str]` or `List[List[str]]`): What the model should recognize in the image. threshold (`float`, *optional*, defaults to 0.1): The probability necessary to make a prediction. top_k (`int`, *optional*, defaults to None): The number of top predictions that will be returned by the pipeline. If the provided number is `None` or higher than the number of predictions available, it will default to the number of predictions. timeout (`float`, *optional*, defaults to None): The maximum time in seconds to wait for fetching images from the web. If None, no timeout is set and the call may block forever. Return: A list of lists containing prediction results, one list per input image. Each list contains dictionaries with the following keys: - **label** (`str`) -- Text query corresponding to the found object. - **score** (`float`) -- Score corresponding to the object (between 0 and 1). - **box** (`Dict[str,int]`) -- Bounding box of the detected object in image's original size. It is a dictionary with `x_min`, `x_max`, `y_min`, `y_max` keys. text_queriesr&r'c3,K|] \}}||dyw)r*Nr).0imglabelss r$ z;ZeroShotObjectDetectionPipeline.__call__..spKCs?ps) pop isinstancestrrlisttuplerr__call__zip)r"r&r'r#inputsresultsr s r$r5z(ZeroShotObjectDetectionPipeline.__call__As~ V #%zz.9  ec5;;/ 0$:JKF e} -,u2E pSVW\^nSop  F'"64V4r%c \i}d|vr|d|d<i}d|vr|d|d<d|vr|d|d<|i|fS)Ntimeout thresholdtop_kr)r"r#preprocess_paramspostprocess_paramss r$_sanitize_parametersz4ZeroShotObjectDetectionPipeline._sanitize_parameterssc  +1)+< i ( & .4[.A { + f *0/ w ' "&888r%c#Kt|d|}|d}t|tr|jd}t j |j |jggtj}t|D]\}}|j||j}|j||j} |jdk(r| j|j} |t|dz k(||d || yw) Nr&)r:r',)dtype)return_tensorsptr )is_last target_sizecandidate_label)rr1r2splittorchtensorheightwidthint32 enumerate tokenizerrimage_processorto torch_dtypelen) r"r7r:r&r'rFirG text_inputsimage_featuress r$ preprocessz*ZeroShotObjectDetectionPipeline.preprocesss6'?G<!"45 & ,/55c: llU\\5;;$?#@ T "+,<"=  A...XK!11%1WN~~%!/!2!243C3C!D$4 5 99*#2  !   sDDc|jd}|jd}|jd}|jdi|}|||d|}|S)NrFrGrE)rFrGrEr)r0model)r" model_inputsrFrGrEoutputs model_outputss r$_forwardz(ZeroShotObjectDetectionPipeline._forwards_"&&}5 &**+<=""9-$**,|,(3dkwovw r%cjg}|D]}|d}t|}|jj|||dd}|djD]I}|d|j } |j |d|d} | || d} |j | Kt|dd }|r|d|}|S) NrGrF)r[r; target_sizesrscoresboxes)scorelabelboxc |dS)Nrbr)xs r$z=ZeroShotObjectDetectionPipeline.postprocess..s ' r%T)keyreverse)rrPpost_process_object_detectionnonzeroitem_get_bounding_boxappendsorted) r"r\r;r<r8 model_outputrcr[indexrbrdresults r$ postprocessz+ZeroShotObjectDetectionPipeline.postprocesss) 'L !23E*<8L**HH$ UbHcIG!*224 ')%0557,,WW-=e-DQ-GH#(5Ev&  ' '&:DI fuoGr%rdz torch.Tensorreturnc|jdk7r td|jj\}}}}||||d}|S)a% Turns list [xmin, xmax, ymin, ymax] into dict { "xmin": xmin, ... } Args: box (`torch.Tensor`): Tensor containing the coordinates in corners format. Returns: bbox (`Dict[str, int]`): Dict containing the coordinates in corners format. rDzAThe ZeroShotObjectDetectionPipeline is only available in PyTorch.)xminyminxmaxymax)rrinttolist)r"rdrvrwrxrybboxs r$rmz1ZeroShotObjectDetectionPipeline._get_bounding_boxsS >>T !`a a!$!1!1!3dD$    r%)N)g?N)__name__ __module__ __qualname____doc__rrr2rrrr5r?rWr]rsrzrm __classcell__)r s@r$rrs@R37VS-d38n)==>V T#Y/Vp 9(,^S#Xr%r)typingrrrrutilsrr r r r baserrPILr image_utilsrrrItransformers.modeling_outputsrmodels.auto.modeling_autor get_loggerr}loggerrrr%r$rsm))kk96=^   H %,FGTmTHTr%