inference_ytvos.py

'''
Inference code for ReferFormer, on Ref-Youtube-VOS
Modified from DETR (https://github.com/facebookresearch/detr)
'''
import argparse
import json
import random
import time
from pathlib import Path

import numpy as np
import torch

import util.misc as utils
from models import build_model
import torchvision.transforms as T
import matplotlib.pyplot as plt
import os
import cv2
from PIL import Image, ImageDraw
import math
import torch.nn.functional as F
import json

import opts
from tqdm import tqdm

import multiprocessing as mp
import threading

from tools.colormap import colormap


# colormap
color_list = colormap()
color_list = color_list.astype('uint8').tolist()

# build transform
transform = T.Compose([
    T.Resize(360),
    T.ToTensor(),
    T.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
    

def main(args):
    args.masks = True
    args.batch_size == 1
    print("Inference only supports for batch size = 1") 

    # fix the seed for reproducibility
    seed = args.seed + utils.get_rank()
    torch.manual_seed(seed)
    np.random.seed(seed)
    random.seed(seed)

    split = args.split
    # save path
    output_dir = args.output_dir
    save_path_prefix = os.path.join(output_dir, split)
    if not os.path.exists(save_path_prefix):
        os.makedirs(save_path_prefix)

    save_visualize_path_prefix = os.path.join(output_dir, split + '_images')
    if args.visualize:
        if not os.path.exists(save_visualize_path_prefix):
            os.makedirs(save_visualize_path_prefix)

    # load data
    root = Path(args.ytvos_path) # data/ref-youtube-vos
    img_folder = os.path.join(root, split, "JPEGImages")
    meta_file = os.path.join(root, "meta_expressions", split, "meta_expressions.json")
    with open(meta_file, "r") as f:
        data = json.load(f)["videos"]
    valid_test_videos = set(data.keys())
    # for some reasons the competition's validation expressions dict contains both the validation (202) & 
    # test videos (305). so we simply load the test expressions dict and use it to filter out the test videos from
    # the validation expressions dict:
    test_meta_file = os.path.join(root, "meta_expressions", "test", "meta_expressions.json")
    with open(test_meta_file, 'r') as f:
        test_data = json.load(f)['videos']
    test_videos = set(test_data.keys())
    valid_videos = valid_test_videos - test_videos
    video_list = sorted([video for video in valid_videos])
    assert len(video_list) == 202, 'error: incorrect number of validation videos'

    # create subprocess
    thread_num = args.ngpu
    global result_dict
    result_dict = mp.Manager().dict()

    processes = []
    lock = threading.Lock()

    video_num = len(video_list)
    per_thread_video_num = video_num // thread_num

    start_time = time.time()
    print('Start inference')
    for i in range(thread_num):
        if i == thread_num - 1:
            sub_video_list = video_list[i * per_thread_video_num:]
        else:
            sub_video_list = video_list[i * per_thread_video_num: (i + 1) * per_thread_video_num]
        p = mp.Process(target=sub_processor, args=(lock, i, args, data, 
                                                   save_path_prefix, save_visualize_path_prefix, 
                                                   img_folder, sub_video_list))
        p.start()
        processes.append(p)

    for p in processes:
        p.join()

    end_time = time.time()
    total_time = end_time - start_time

    result_dict = dict(result_dict)
    num_all_frames_gpus = 0
    for pid, num_all_frames in result_dict.items():
        num_all_frames_gpus += num_all_frames

    print("Total inference time: %.4f s" %(total_time))

def sub_processor(lock, pid, args, data, save_path_prefix, save_visualize_path_prefix, img_folder, video_list):
    text = 'processor %d' % pid
    with lock:
        progress = tqdm(
            total=len(video_list),
            position=pid,
            desc=text,
            ncols=0
        )
    torch.cuda.set_device(pid)

    # model
    model, criterion, _ = build_model(args) 
    device = args.device
    model.to(device)

    model_without_ddp = model
    n_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad)

    if pid == 0:
        print('number of params:', n_parameters)

    if args.resume:
        checkpoint = torch.load(args.resume, map_location='cpu')
        missing_keys, unexpected_keys = model_without_ddp.load_state_dict(checkpoint['model'], strict=False)
        unexpected_keys = [k for k in unexpected_keys if not (k.endswith('total_params') or k.endswith('total_ops'))]
        if len(missing_keys) > 0:
            print('Missing Keys: {}'.format(missing_keys))
        if len(unexpected_keys) > 0:
            print('Unexpected Keys: {}'.format(unexpected_keys))
    else:
    	raise ValueError('Please specify the checkpoint for inference.')


    # start inference
    num_all_frames = 0 
    model.eval()

    # 1. For each video
    for video in video_list:
        metas = [] # list[dict], length is number of expressions

        expressions = data[video]["expressions"]   
        expression_list = list(expressions.keys()) 
        num_expressions = len(expression_list)
        video_len = len(data[video]["frames"])

        # read all the anno meta
        for i in range(num_expressions):
            meta = {}
            meta["video"] = video
            meta["exp"] = expressions[expression_list[i]]["exp"]
            meta["exp_id"] = expression_list[i]
            meta["frames"] = data[video]["frames"]
            metas.append(meta)
        meta = metas

        # 2. For each expression
        for i in range(num_expressions):
            video_name = meta[i]["video"]
            exp = meta[i]["exp"]
            exp_id = meta[i]["exp_id"]
            frames = meta[i]["frames"]

            video_len = len(frames)
            # store images
            imgs = []
            for t in range(video_len):
                frame = frames[t]
                img_path = os.path.join(img_folder, video_name, frame + ".jpg")
                img = Image.open(img_path).convert('RGB')
                origin_w, origin_h = img.size
                imgs.append(transform(img)) # list[img]

            imgs = torch.stack(imgs, dim=0).to(args.device) # [video_len, 3, h, w]
            img_h, img_w = imgs.shape[-2:]
            size = torch.as_tensor([int(img_h), int(img_w)]).to(args.device)
            target = {"size": size}

            with torch.no_grad():
                outputs = model([imgs], [exp], [target])
            
            pred_logits = outputs["pred_logits"][0] 
            pred_boxes = outputs["pred_boxes"][0]   
            pred_masks = outputs["pred_masks"][0]   
            pred_ref_points = outputs["reference_points"][0]  

            # according to pred_logits, select the query index
            pred_scores = pred_logits.sigmoid() # [t, q, k]
            pred_scores = pred_scores.mean(0)   # [q, k]
            max_scores, _ = pred_scores.max(-1) # [q,]
            _, max_ind = max_scores.max(-1)     # [1,]
            max_inds = max_ind.repeat(video_len)
            pred_masks = pred_masks[range(video_len), max_inds, ...] # [t, h, w]
            pred_masks = pred_masks.unsqueeze(0)

            pred_masks = F.interpolate(pred_masks, size=(origin_h, origin_w), mode='bilinear', align_corners=False) 
            pred_masks = (pred_masks.sigmoid() > args.threshold).squeeze(0).detach().cpu().numpy() 

            # store the video results
            all_pred_logits = pred_logits[range(video_len), max_inds] 
            all_pred_boxes = pred_boxes[range(video_len), max_inds]   
            all_pred_ref_points = pred_ref_points[range(video_len), max_inds] 
            all_pred_masks = pred_masks

            if args.visualize:
                for t, frame in enumerate(frames):
                    # original
                    img_path = os.path.join(img_folder, video_name, frame + '.jpg')
                    source_img = Image.open(img_path).convert('RGBA') # PIL image

                    draw = ImageDraw.Draw(source_img)
                    draw_boxes = all_pred_boxes[t].unsqueeze(0) 
                    draw_boxes = rescale_bboxes(draw_boxes.detach(), (origin_w, origin_h)).tolist()

                    # draw boxes
                    xmin, ymin, xmax, ymax = draw_boxes[0]
                    draw.rectangle(((xmin, ymin), (xmax, ymax)), outline=tuple(color_list[i%len(color_list)]), width=2)

                    # draw reference point
                    ref_points = all_pred_ref_points[t].unsqueeze(0).detach().cpu().tolist() 
                    draw_reference_points(draw, ref_points, source_img.size, color=color_list[i%len(color_list)])

                    # draw mask
                    source_img = vis_add_mask(source_img, all_pred_masks[t], color_list[i%len(color_list)])

                    # save
                    save_visualize_path_dir = os.path.join(save_visualize_path_prefix, video, str(i))
                    if not os.path.exists(save_visualize_path_dir):
                        os.makedirs(save_visualize_path_dir)
                    save_visualize_path = os.path.join(save_visualize_path_dir, frame + '.png')
                    source_img.save(save_visualize_path)


            # save binary image
            save_path = os.path.join(save_path_prefix, video_name, exp_id)
            if not os.path.exists(save_path):
                os.makedirs(save_path)
            for j in range(video_len):
                frame_name = frames[j]
                mask = all_pred_masks[j].astype(np.float32) 
                mask = Image.fromarray(mask * 255).convert('L')
                save_file = os.path.join(save_path, frame_name + ".png")
                mask.save(save_file)

        with lock:
            progress.update(1)
    result_dict[str(pid)] = num_all_frames
    with lock:
        progress.close()


# visuaize functions
def box_cxcywh_to_xyxy(x):
    x_c, y_c, w, h = x.unbind(1)
    b = [(x_c - 0.5 * w), (y_c - 0.5 * h),
         (x_c + 0.5 * w), (y_c + 0.5 * h)]
    return torch.stack(b, dim=1)

def rescale_bboxes(out_bbox, size):
    img_w, img_h = size
    b = box_cxcywh_to_xyxy(out_bbox)
    b = b.cpu() * torch.tensor([img_w, img_h, img_w, img_h], dtype=torch.float32)
    return b


# Visualization functions
def draw_reference_points(draw, reference_points, img_size, color):
    W, H = img_size
    for i, ref_point in enumerate(reference_points):
        init_x, init_y = ref_point
        x, y = W * init_x, H * init_y
        cur_color = color
        draw.line((x-10, y, x+10, y), tuple(cur_color), width=4)
        draw.line((x, y-10, x, y+10), tuple(cur_color), width=4)

def draw_sample_points(draw, sample_points, img_size, color_list):
    alpha = 255
    for i, samples in enumerate(sample_points):
        for sample in samples:
            x, y = sample
            cur_color = color_list[i % len(color_list)][::-1]
            cur_color += [alpha]
            draw.ellipse((x-2, y-2, x+2, y+2), 
                            fill=tuple(cur_color), outline=tuple(cur_color), width=1)

def vis_add_mask(img, mask, color):
    origin_img = np.asarray(img.convert('RGB')).copy()
    color = np.array(color)

    mask = mask.reshape(mask.shape[0], mask.shape[1]).astype('uint8') # np
    mask = mask > 0.5

    origin_img[mask] = origin_img[mask] * 0.5 + color * 0.5
    origin_img = Image.fromarray(origin_img)
    return origin_img

  

if __name__ == '__main__':
    parser = argparse.ArgumentParser('ReferFormer inference script', parents=[opts.get_args_parser()])
    args = parser.parse_args()
    main(args)