从光流估计网络输出光流
光流估计网络
LiteFlowNet2
GeoNet 光流深度位姿联合估计
Nvidia flownet2 pytorch
PWC-Net
光流保存为 png,jpg,flo 文件
pre_flow = model(data )
# save flow
flow_fn = '%.6d.png' % tgt_idx
color_fn = os.path.join(color_dir, flow_fn)
color_flow = fl.flow_to_image(pred_flow)
color_flow = cv2.cvtColor(color_flow, cv2.COLOR_RGB2BGR)
color_flow = cv2.imwrite(color_fn, color_flow)
png_fn = os.path.join(png_dir, flow_fn)
mask_blob = np.ones((opt.img_height, opt.img_width), dtype = np.uint16)
fl.write_kitti_png_file(png_fn, pred_flow, mask_blob)
binary_fn = flow_fn.replace('.png', '.flo')
binary_fn = os.path.join(binary_dir, binary_fn)
fl.write_flow(pred_flow, binary_fn)