Duplicate from lei-qi-233/MicroG-4M

Co-authored-by: Lei Qi <lei-qi-233@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,61 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mds filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text
+# Video files - compressed
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text
+csv_files/bounding_boxes.csv filter=lfs diff=lfs merge=lfs -text
+annotation_files/bounding_boxes.csv filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,207 @@
+---
+license: cc-by-4.0
+
+configs:
+- config_name: actions
+  data_files:
+  - split: full
+    path:
+    - annotation_files/actions.csv
+  format: csv
+- config_name: bounding_boxes
+  data_files:
+  - split: full
+    path:
+    - annotation_files/bounding_boxes.csv
+  format: csv
+- config_name: vqa_and_captioning
+  data_files:
+  - split: full
+    path:
+    - vqa_and_captioning/vqa_and_captioning.json
+  format: json
+language:
+- en
+tags:
+- human action recognition
+- video understanding
+- microgravity
+task_categories:
+- video-classification
+- visual-question-answering
+- video-text-to-text
+
+---
+
+# MicroG-4M Dataset
+
+This repository stores the entire content of the  MicroG-4M dataset itself.
+
+**For more information and details, including training, evaluation, statistics, and related code, please:**
+
+- **Refer to our [paper](https://arxiv.org/abs/2506.02845)**
+
+- **Visit our [GitHub](https://github.com/LEI-QI-233/MicroG-4M)**
+
+- **And check our [fine-tuned models](https://huggingface.co/LEI-QI-233/MicroG-4M-models)**
+
+
+
+
+# Specification of MicroG-4M
+
+## "annotation_files" Folder
+
+The folder contains all annotation files of the dataset, all stored in CSV format.
+
+### [actions.csv](./annotation_files/actions.csv)
+
+contains all the labeled persons in the video clips and their corresponding action IDs.
+
+The headers are:
+- `video_id` :  is the [name](#video-name) of video clip.
+- `movie_or_real`: `m` for movie and `r` for real footage
+- `person_id`: corresponds to the person labeled by the respective bounding box. You can visually identify the corresponding person directly in the videos within the [videos_annotated](./videos_annotated.zip) folder.
+- `action`: the ID of action. For the mapping relationship between action IDs and labels see [label_map](#label_map-folder)
+
+### [bounding_boxes.csv](./annotation_files/bounding_boxes.csv)
+
+includes all detected persons and their corresponding bounding box coordinates. The coordinates are expressed in pixels rather than normalized values.
+
+
+The headers are:
+- `video_id`: is the [name](#video-name) of video clip.
+- `frame_id`: Not a continuous number. Only frames where the bounding box is drawn. If no person is detected, ignore this frame.
+- `person_id`
+- `xmin`: x coordinate of the upper left corner
+
+- `ymin`: y coordinate of the upper left corner
+
+- `xmax`: x coordinate of the lower right corner
+
+- `ymax`: y coordinate of the lower right corner
+
+
+### NOTE
+
+Each row in the [actions.csv](./annotation_files/actions.csv) and [bounding_boxes.csv](./annotation_files/bounding_boxes.csv) files is associated through the video ID and person ID, forming a complete information structure that includes: 
+- video ID
+- frame ID
+- person ID
+- movie or real
+- action ID
+- bounding box coordinates
+
+## "label_map" Folder
+The folder contains the pbtxt file of the label map as well as a PDF file.
+
+### [label_map.pbtxt](./label_map/label_map.pbtxt)
+is the pbtxt format file of the label map, containing three components: id, name, and type. Its format is consistent with the label map format of the [AVA](https://research.google.com/ava/) dataset. If this dataset is converted to the AVA format for training or evaluation, this label map file can be directly used as input.
+
+### [label_map.pdf](./label_map/label_map.pdf)
+ is the PDF format of the label map, providing a convenient reference for users to read and consult.
+
+
+## "vqa_and_captioning" Folder
+The folder contains the json files for Visual Queastion Qnswering and captioning.
+
+### [vqa_and_captioning.json](./vqa_and_captioning/vqa_and_captioning.json)
+is an array containing information for multiple videos, with each record consisting of the following parts:
+
+- `video_id`: The ID of the video
+- `caption`: A brief description of the video content, used for generating subtitles
+- `qa_pairs`: A list of question-answer pairs, where each object includes two fields:
+    - `Q`: A question related to the video scene or context
+    - `A`: The corresponding answer to the question
+
+<!--## [videos.zip](videos.zip)
+The compressed package is a folder containing all video clips of the dataset. 
+
+All video clips are valid clips and completely correspond to the files in "annotation_files" folder.
+
+ -->
+
+## "videos" Folder
+The [`video_id_list.pdf`](./videos/video_id_list.pdf) file in this folder provides the identifiers for all videos used in the dataset. **Users should download the raw files independently** and process them into the required format as described in our paper.
+
+You can process the downloaded videos into our dataset format using your own tools or our provided automated pipeline, please refer to [`CODE.md`](https://github.com/lei-qi-233/MicroG-4M/blob/main/code/CODE.md) in our GitHub repository.
+
+## Video Specification 
+
+### Format of processed videos
+All videos are processed to video clips which are 3 seconds long, with a frame rate of 30 fps and encoded in H.264 format. You can find detailed information in our Github and our paper.
+
+### Video Folder Structure
+
+The video folder should has the following structure:
+
+```
+videos
+ |_ movie
+ |  |_ [movie name 0]
+ |  |  |_ [movie name 0]_000.mp4
+ |  |  |_ [movie name 0]_003.mp4
+ |  |  |_ ...
+ |  |_ [movie name 1]
+ |  |  |_ ...
+ |_ real
+ |  |_ [real video id 0]
+ |  |  |_[real video id 0]_002.mp4
+ |  |  |_[real video id 0]_003.mp4
+ |  |  |_ ...
+ |  |_ [real video id 1]
+ |  |  |_ ...
+```
+
+
+#### Video Name
+The video name consists of a sequence of characters and underscores followed by a sequence of numbers, and corresponds to the `video_id` column in the CSV and JSON annotation files. The character string is either the YouTube video ID or the film name. 
+
+The sequence number at the end of the video name indicates its specific position within the original video. For example, 000 represents the 0 to 3 seconds of the original video, and 003 represents the 9 to 12 seconds. Since not all clips meet the criteria, the sequence numbers of video clips in the dataset are not consecutive.
+
+For example:
+
+- `AErpXJq67LM_000` represents the first three-second segment of the YouTube video with ID `AErpXJq67LM`
+- `Apollo13_101` denotes the segment of the movie `Apollo13` spanning the 303rd to the 306rd second (i.e., from number × 3 to (number × 3) + 3 seconds).
+
+#### `movie` and `real` folders
+- In the `movie` folder, all video clips are from movies.  `movie name` is the name of the movie. 
+
+- In the `real` folder, all videos are real footage and downloaded from YouTube. `real video id` is the video id on YouTube. 
+
+### For further information, please refer to our paper.
+
+
+
+<!--## [videos_annotated.zip](./videos_annotated.zip)
+The contents and structure of the `videos_annotated` folder are identical to those of the `videos` folder. The only difference is that the videos in the `videos_annotated` folder include bounding boxes to annotate detected person frames and their corresponding action IDs, consistent with the [bounding_boxes.csv](./annotation_files/bounding_boxes.csv) file, facilitating manual review and validation.
+-->
+<!-- ## "parquet_format" Folder
+This folder contains all the data files in Parquet format, which are used by Hugging Face to automatically generate the Croissant file.
+
+The Croissant JSON-LD metadata references these Parquet files. When you load the dataset via the Croissant JSON-LD, it will download these Parquet files, not original csv and json files.
+
+### [actions.parquet](./parquet_format/actions.parquet)
+is the parquet format of [actions.csv](./annotation_files/actions.csv)
+
+### [bounding_boxes.parquet](./parquet_format/bounding_boxes.parquet)
+is the parquet format of [bounding_boxes.csv](./annotation_files/bounding_boxes.csv)
+
+### [vqa_and_captioning.parquet](./parquet_format/vqa_and_captioning.parquet)
+is the parquet format of [vqa_and_captioning.json](./vqa_and_captioning/vqa_and_captioning.json)
+
+### NOTE
+The parquet files do not include <!--video clips ([videos.zip](videos.zip)) and label map ([label_map.pbtxt](./label_map/label_map.pbtxt)). You can check or download it from this repository if you need. -->
+
+# Citing MicroG-4M
+If you find this dataset useful in your research, please use the following BibTeX entry for citation:
+
+```bibtex
+@inproceedings{wen2026go,
+  title={{Go Beyond Earth: Understanding Human Actions and Scenes in Microgravity Environments}},
+  author={Di Wen and Lei Qi and Kunyu Peng and Kailun Yang and Fei Teng and Ao Luo and Jia Fu and Yufan Chen and Ruiping Liu and Yitian Shi and M. Saquib Sarfraz and Rainer Stiefelhagen},
+  booktitle={International Conference on Learning Representations (ICLR)},
+  year={2026},
+  url={https://openreview.net/forum?id=gygGCVXeh3}
+}
+```

annotation_files/bounding_boxes.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09f20c064715dc81fef93aab58c5f8d8034d20daf5aeaaf74b2e349d1b2a9adc
+size 13433864

label_map/label_map.pbtxt

@@ -0,0 +1,299 @@
+item {
+  name: "bend/bow (at the waist)"
+  id: 1
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "crouch/kneel"
+  id: 3
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "fall down"
+  id: 5
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "get up"
+  id: 6
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "jump/leap"
+  id: 7
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "lie/sleep"
+  id: 8
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "martial art"
+  id: 9
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "run/jog"
+  id: 10
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "sit"
+  id: 11
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "stand"
+  id: 12
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "walk"
+  id: 14
+  label_type: PERSON_MOVEMENT
+}
+
+item {
+  name: "carry/hold (an object)"
+  id: 17
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "climb (e.g., a mountain)"
+  id: 20
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "close (e.g., a door, a box)"
+  id: 22
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "cut"
+  id: 24
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "dress/undress clothing"
+  id: 26
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "drink"
+  id: 27
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "operate spaceship"
+  id: 28
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "eat"
+  id: 29
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "enter"
+  id: 30
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "hit (an object)"
+  id: 34
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "lift/pick up"
+  id: 36
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "open (e.g., a window, a car door)"
+  id: 38
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "play musical instrument"
+  id: 41
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "point to (an object)"
+  id: 43
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "pull (an object)"
+  id: 45
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "push (an object)"
+  id: 46
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "put down"
+  id: 47
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "read"
+  id: 48
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "take a photo"
+  id: 56
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "text on/look at a cellphone"
+  id: 57
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "throw"
+  id: 58
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "touch (an object)"
+  id: 59
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "turn (e.g., a screwdriver)"
+  id: 60
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "watch (e.g., TV)/any unspecified action"
+  id: 61
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "work on a computer"
+  id: 62
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "write"
+  id: 63
+  label_type: OBJECT_MANIPULATION
+}
+
+item {
+  name: "fight/hit (a person)"
+  id: 64
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "give/serve (an object) to (a person)"
+  id: 65
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "grab (a person)"
+  id: 66
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "hand clap"
+  id: 67
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "hand shake"
+  id: 68
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "hand wave"
+  id: 69
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "hug (a person)"
+  id: 70
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "kiss (a person)"
+  id: 72
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "listen to (a person)"
+  id: 74
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "push (another person)"
+  id: 76
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "take (an object) from (a person)"
+  id: 78
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "talk to (e.g., self, a person, a group)"
+  id: 79
+  label_type: PERSON_INTERACTION
+}
+
+item {
+  name: "watch (a person)"
+  id: 80
+  label_type: PERSON_INTERACTION
+}

label_map/label_map.pdf

@@ -0,0 +1,74 @@
+%PDF-1.4
+%���� ReportLab Generated PDF document http://www.reportlab.com
+1 0 obj
+<<
+/F1 2 0 R /F2 3 0 R
+>>
+endobj
+2 0 obj
+<<
+/BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font
+>>
+endobj
+3 0 obj
+<<
+/BaseFont /Helvetica-Bold /Encoding /WinAnsiEncoding /Name /F2 /Subtype /Type1 /Type /Font
+>>
+endobj
+4 0 obj
+<<
+/Contents 8 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 7 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+5 0 obj
+<<
+/PageMode /UseNone /Pages 7 0 R /Type /Catalog
+>>
+endobj
+6 0 obj
+<<
+/Author (\(anonymous\)) /CreationDate (D:20250502204424-01'00') /Creator (\(unspecified\)) /Keywords () /ModDate (D:20250502204424-01'00') /Producer (ReportLab PDF Library - www.reportlab.com) 
+  /Subject (\(unspecified\)) /Title (\(anonymous\)) /Trapped /False
+>>
+endobj
+7 0 obj
+<<
+/Count 1 /Kids [ 4 0 R ] /Type /Pages
+>>
+endobj
+8 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 1964
+>>
+stream
+GatU5gN)%,&:O:Si6fZR&cdL`RC@:d90,Z!,appmKq]]H2FXpa=NTG:+>2Yc$pi@&a.YHTBQ:87>lZ'7=T+?WhsuHI:E-\?Kci_0Ard4%0spb,IbrlYF%7c;k:=E*8<CpaPor1hT2RG8iVmeo4:ps\TZp%gUkki#+f3m5+_69edJ.=/mXKT\)-%.BdM8U+K0BuR=eh$HpBkaYdH3=_&^lr1:[7*h4dkec!!,/J1?6Ra?=d*?Pb@K;C'or$:)IFi!-q=rd_J[4dJc"jjr>d3#E%4h/LI+J'h]/jOMH^5_Ys@UDG$!]eL1aLM'p"ha/X"'+2pgcr!AC.QHrP$1-tD0`jl`LG[&kff%]nW*q"puNCSiN75^#7YCZk4FT,Y\)o2$e^7EF,qPfU_%f56sk,'l*QFAO4)XZ7"'6M=Xol\Sj%P@,l7l5?SDi"e8Lj?>>a6D*UCKZ<2&ek0ok0W\a9!3qQb-46LJt9q%>2H7Ba/Z)#HA#P#\VW2[&YG*9:X/a@4Tt]i'5A.aR>s7+@;Xd&Ra@ET9,tsGf`f&D>;@blbnA>kr=j@>QjAo.@WB>lkkf;a@pADD#r\a<50EFgrMA"sRn!!YM.Kt`[`Z'a*EGCt!hJ&E)2kN*1'BKQT[E74@Tg0m>R[lQgKm+RmFUhU/2t,5Z*6<!3<^3%'.,Ll<q)-ZlfBh%Ve1B%V11U.9lTJTDK80U8ThsW?T#YGUA'i(>HI`CZ0c(/N"QSG,92_DYRRhT-8*jS*)RL1OAZbb,dtu<;)Y39KANhhHP0&],p',W[X(Jb]-#!imtc+o+4jdu5$dFJ8S00=`)>ZP.RhUV!7"T>[O;$?.dsnfY"XA8>d.=?FMt=7^7j"4<2Jm<<ATU4FAfpcqG0h7=DZ$$`3-3\FVJM-N.5:L`aYFH,Bf:)]Phjc9u^9l[?fH>_sp4J`8los.bYKHBtScOTXi<VCrJAC08nbK6o1$`1cB>D*/c2D(0c!!a#<.GS+rnd.6HN;^Q,70eof_2+a#bDMcnW=)P[V15&"aEO[c*RM;tjZcTuKn*XcKFbRm?ne@]W!:hid^Z1d=9j4#_4Hmj>e<c.h=?7B_`9R:rbZlS[AN:R1r.%@sn?EK.J)k*4hP@_/*j:2pr*?[0_a\D2JD!Q8_F5&pY,jXO"-ah%5/Y$>uAN\O6*:_fc@rS7U@8b*T)hK=&X2Ap^gMHtj*H>br*+;BQdi7UFNd+Bg>S",iQ>lfT4r>B'fi6!aee&F@Z+\C<[4.*cXWnn;=]ZVX_Uf;bCaNXllA2deiM`2=cm?-bN0UCfc7[G^F3nE/nfhQ9BmIb`E)STI]un=Z_t'BKFZd=UDe``7U3?leUDl[F@^pDt0p;eR4l(Rr%B<@/m%`DNP7\Bil]B:r")DH)\#8!3Z;#(UQ.V\//m9SkVBHGThWs<>L=(JkqKD/,9[-hK@qu?.h\W2a^GYjnrBB)?;G;++d.bl4^YZ(*FS=T(SM&e4BsU99AdC$j%/q2!L^QhjOh_OF(M>kj91YNn&emURoMCtW8=:N.?p<<::]b0@dmF8\+:B&(b.nf2Ba/*]dmF2Xd&)RXOJ,AZ0d&iQ*_LqkD0h"').Xlm6Q,YCKB5@Q2E`=9$uL<H!u]hb&-F\pl(J7$d&)RHPX-*@0U.,A16NWud&)R$OJ,AZE?IW<*_LtlD0dT21:69h$>%S<$k-_*M[Z$V);=^Z#%VDr6QF1oVMA%l6^o,hJ.-:@LdDu$+bN#tj<?rKO<p\s%qS"\2Ef!0$uL;="Z!IL+U2IjdmFhlTF2V?arhK<3DbiF6tAPM[VT6RF"OC/(@bj"_tC[`/PPP@RW+Q.&D6](/\rC/YC6hl6K$koL>YKeRP%qj]_p'-RWfhBh8&tro4EV=[a.iYj56O(mfl0$.-9@paac(2r7>rgS9JoDZ"PV0,qnnm_N$5?i8BtE(d&uAF40OMG=L^aA&dY)Nt"!`4Z9nK(CHN%J,~>endstream
+endobj
+xref
+0 9
+0000000000 65535 f 
+0000000073 00000 n 
+0000000114 00000 n 
+0000000221 00000 n 
+0000000333 00000 n 
+0000000536 00000 n 
+0000000604 00000 n 
+0000000887 00000 n 
+0000000946 00000 n 
+trailer
+<<
+/ID 
+[<b4fcf6b9855f0bc49a935a45e53b101c><b4fcf6b9855f0bc49a935a45e53b101c>]
+% ReportLab generated PDF document -- digest (http://www.reportlab.com)
+
+/Info 6 0 R
+/Root 5 0 R
+/Size 9
+>>
+startxref
+3001
+%%EOF