195 lines
12 KiB
Markdown
195 lines
12 KiB
Markdown
---
|
||
layout: default
|
||
title: Objectron (3D Object Detection)
|
||
parent: Solutions
|
||
nav_order: 10
|
||
---
|
||
|
||
# MediaPipe Objectron
|
||
{: .no_toc }
|
||
|
||
1. TOC
|
||
{:toc}
|
||
---
|
||
|
||
## Overview
|
||
|
||
MediaPipe Objectron is a mobile real-time 3D object detection solution for
|
||
everyday objects. It detects objects in 2D images, and estimates their poses and
|
||
sizes through a machine learning (ML) model, trained on a newly created 3D
|
||
dataset.
|
||
|
||
 | 
|
||
:--------------------------------------------------------------------------------: | :----------------------------------------------------------------------------------:
|
||
*Fig 1(a). Objectron for Shoes.* | *Fig 1(b). Objectron for Chairs.*
|
||
|
||
Object detection is an extensively studied computer vision problem, but most of
|
||
the research has focused on
|
||
[2D object prediction](https://ai.googleblog.com/2017/06/supercharge-your-computer-vision-models.html).
|
||
While 2D prediction only provides 2D bounding boxes, by extending prediction to
|
||
3D, one can capture an object’s size, position and orientation in the world,
|
||
leading to a variety of applications in robotics, self-driving vehicles, image
|
||
retrieval, and augmented reality. Although 2D object detection is relatively
|
||
mature and has been widely used in the industry, 3D object detection from 2D
|
||
imagery is a challenging problem, due to the lack of data and diversity of
|
||
appearances and shapes of objects within a category.
|
||
|
||
 |
|
||
:-----------------------------------------------------------------------: |
|
||
*Fig 2. Objectron example results.* |
|
||
|
||
## Obtaining Real-World 3D Training Data
|
||
|
||
While there are ample amounts of 3D data for street scenes, due to the
|
||
popularity of research into self-driving cars that rely on 3D capture sensors
|
||
like LIDAR, datasets with ground truth 3D annotations for more granular everyday
|
||
objects are extremely limited. To overcome this problem, we developed a novel
|
||
data pipeline using mobile augmented reality (AR) session data. With the arrival
|
||
of [ARCore](https://developers.google.com/ar) and
|
||
[ARKit](https://developer.apple.com/augmented-reality/),
|
||
[hundreds of millions](https://arinsider.co/2019/05/13/arcore-reaches-400-million-devices/)
|
||
of smartphones now have AR capabilities and the ability to capture additional
|
||
information during an AR session, including the camera pose, sparse 3D point
|
||
clouds, estimated lighting, and planar surfaces.
|
||
|
||
In order to label ground truth data, we built a novel annotation tool for use
|
||
with AR session data, which allows annotators to quickly label 3D bounding boxes
|
||
for objects. This tool uses a split-screen view to display 2D video frames on
|
||
which are overlaid 3D bounding boxes on the left, alongside a view showing 3D
|
||
point clouds, camera positions and detected planes on the right. Annotators draw
|
||
3D bounding boxes in the 3D view, and verify its location by reviewing the
|
||
projections in 2D video frames. For static objects, we only need to annotate an
|
||
object in a single frame and propagate its location to all frames using the
|
||
ground truth camera pose information from the AR session data, which makes the
|
||
procedure highly efficient.
|
||
|
||
|  |
|
||
| :--------------------------------------------------------------------------: |
|
||
| *Fig 3. Real-world data annotation for 3D object detection. (Right) 3D bounding boxes are annotated in the 3D world with detected surfaces and point clouds. (Left) Projections of annotated 3D bounding boxes are overlaid on top of video frames making it easy to validate the annotation.* |
|
||
|
||
## AR Synthetic Data Generation
|
||
|
||
A popular approach is to complement real-world data with synthetic data in order
|
||
to increase the accuracy of prediction. However, attempts to do so often yield
|
||
poor, unrealistic data or, in the case of photorealistic rendering, require
|
||
significant effort and compute. Our novel approach, called AR Synthetic Data
|
||
Generation, places virtual objects into scenes that have AR session data, which
|
||
allows us to leverage camera poses, detected planar surfaces, and estimated
|
||
lighting to generate placements that are physically probable and with lighting
|
||
that matches the scene. This approach results in high-quality synthetic data
|
||
with rendered objects that respect the scene geometry and fit seamlessly into
|
||
real backgrounds. By combining real-world data and AR synthetic data, we are
|
||
able to increase the accuracy by about 10%.
|
||
|
||
 |
|
||
:-------------------------------------------------------------------------------------------: |
|
||
*Fig 4. An example of AR synthetic data generation. The virtual white-brown cereal box is rendered into the real scene, next to the real blue book.* |
|
||
|
||
## ML Model for 3D Object Detection
|
||
|
||
 |
|
||
:---------------------------------------------------------------------------------: |
|
||
*Fig 5. Network architecture and post-processing for 3D object detection.* |
|
||
|
||
We [built a single-stage model](https://arxiv.org/abs/2003.03522) to predict the
|
||
pose and physical size of an object from a single RGB image. The model backbone
|
||
has an encoder-decoder architecture, built upon
|
||
[MobileNetv2](https://ai.googleblog.com/2018/04/mobilenetv2-next-generation-of-on.html).
|
||
We employ a multi-task learning approach, jointly predicting an object's shape
|
||
with detection and regression. The shape task predicts the object's shape
|
||
signals depending on what ground truth annotation is available, e.g.
|
||
segmentation. This is optional if there is no shape annotation in training data.
|
||
For the detection task, we use the annotated bounding boxes and fit a Gaussian
|
||
to the box, with center at the box centroid, and standard deviations
|
||
proportional to the box size. The goal for detection is then to predict this
|
||
distribution with its peak representing the object’s center location. The
|
||
regression task estimates the 2D projections of the eight bounding box vertices.
|
||
To obtain the final 3D coordinates for the bounding box, we leverage a well
|
||
established pose estimation algorithm
|
||
([EPnP](https://www.epfl.ch/labs/cvlab/software/multi-view-stereo/epnp/)). It
|
||
can recover the 3D bounding box of an object, without a priori knowledge of the
|
||
object dimensions. Given the 3D bounding box, we can easily compute pose and
|
||
size of the object. The model is light enough to run real-time on mobile devices
|
||
(at 26 FPS on an Adreno 650 mobile GPU).
|
||
|
||
 |
|
||
:-------------------------------------------------------------------------------------: |
|
||
*Fig 6. Sample results of our network — (Left) original 2D image with estimated bounding boxes, (Middle) object detection by Gaussian distribution, (Right) predicted segmentation mask.* |
|
||
|
||
## Detection and Tracking Pipeline
|
||
|
||
When the model is applied to every frame captured by the mobile device, it can
|
||
suffer from jitter due to the ambiguity of the 3D bounding box estimated in each
|
||
frame. To mitigate this, we adopt the same detection+tracking strategy in our
|
||
[2D object detection and tracking pipeline](./box_tracking.md#object-detection-and-tracking)
|
||
in [MediaPipe Box Tracking](./box_tracking.md). This mitigates the need to run
|
||
the network on every frame, allowing the use of heavier and therefore more
|
||
accurate models, while keeping the pipeline real-time on mobile devices. It also
|
||
retains object identity across frames and ensures that the prediction is
|
||
temporally consistent, reducing the jitter.
|
||
|
||
The Objectron 3D object detection and tracking pipeline is implemented as a
|
||
MediaPipe
|
||
[graph](https://github.com/google/mediapipe/tree/master/mediapipe/graphs/object_detection_3d/shoe_classic_occlusion_tracking.pbtxt),
|
||
which internally uses a
|
||
[detection subgraph](https://github.com/google/mediapipe/tree/master/mediapipe/graphs/object_detection_3d/subgraphs/objectron_detection_gpu.pbtxt)
|
||
and a
|
||
[tracking subgraph](https://github.com/google/mediapipe/tree/master/mediapipe/graphs/object_detection_3d/subgraphs/objectron_tracking_gpu.pbtxt).
|
||
The detection subgraph performs ML inference only once every few frames to
|
||
reduce computation load, and decodes the output tensor to a FrameAnnotation that
|
||
contains nine keypoints: the 3D bounding box's center and its eight vertices.
|
||
The tracking subgraph runs every frame, using the box traker in
|
||
[MediaPipe Box Tracking](./box_tracking.md) to track the 2D box tightly
|
||
enclosing the projection of the 3D bounding box, and lifts the tracked 2D
|
||
keypoints to 3D with
|
||
[EPnP](https://www.epfl.ch/labs/cvlab/software/multi-view-stereo/epnp/). When
|
||
new detection becomes available from the detection subgraph, the tracking
|
||
subgraph is also responsible for consolidation between the detection and
|
||
tracking results, based on the area of overlap.
|
||
|
||
## Example Apps
|
||
|
||
Please first see general instructions for
|
||
[Android](../getting_started/building_examples.md#android) and
|
||
[iOS](../getting_started/building_examples.md#ios) on how to build MediaPipe examples.
|
||
|
||
Note: To visualize a graph, copy the graph and paste it into
|
||
[MediaPipe Visualizer](https://viz.mediapipe.dev/). For more information on how
|
||
to visualize its associated subgraphs, please see
|
||
[visualizer documentation](../tools/visualizer.md).
|
||
|
||
### Objectron for Shoes
|
||
|
||
* Graph:
|
||
[`mediapipe/graphs/object_detection_3d/shoe_classic_occlusion_tracking.pbtxt`](https://github.com/google/mediapipe/tree/master/mediapipe/graphs/object_detection_3d/shoe_classic_occlusion_tracking.pbtxt)
|
||
* Android target:
|
||
[(or download prebuilt ARM64 APK)](https://drive.google.com/open?id=1S0K4hbWt3o31FfQ4QU3Rz7IHrvOUMx1d)
|
||
[`mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetection3d:objectdetection3d`](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetection3d/BUILD)
|
||
* iOS target: Not available
|
||
|
||
### Objectron for Chairs
|
||
|
||
* Graph:
|
||
[`mediapipe/graphs/hair_segmentation/hair_segmentation_mobile_gpu.pbtxt`](https://github.com/google/mediapipe/tree/master/mediapipe/graphs/object_detection_3d/chair_classic_occlusion_tracking.pbtxt)
|
||
* Android target:
|
||
[(or download prebuilt ARM64 APK)](https://drive.google.com/open?id=1MM8K-13bXLCVS1EHQ-KgkVyEahEPrKej)
|
||
[`mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetection3d:objectdetection3d`](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetection3d/BUILD)
|
||
and add `--define chair=true` to the build command, i.e.,
|
||
|
||
```bash
|
||
bazel build -c opt --config android_arm64 --define chair=true mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetection3d:objectdetection3d
|
||
```
|
||
|
||
* iOS target: Not available
|
||
|
||
## Resources
|
||
|
||
* Google AI Blog:
|
||
[Real-Time 3D Object Detection on Mobile Devices with MediaPipe](https://ai.googleblog.com/2020/03/real-time-3d-object-detection-on-mobile.html)
|
||
* Paper: [MobilePose: Real-Time Pose Estimation for Unseen Objects with Weak
|
||
Shape Supervision](https://arxiv.org/abs/2003.03522)
|
||
* Paper:
|
||
[Instant 3D Object Tracking with Applications in Augmented Reality](https://drive.google.com/open?id=1O_zHmlgXIzAdKljp20U_JUkEHOGG52R8)
|
||
([presentation](https://www.youtube.com/watch?v=9ndF1AIo7h0))
|
||
* [Models and model cards](./models.md#objectron)
|