1. Information

• Submission Window: From 11:59 PM Pacific Time on March 1, 2025, to 11:59 PM Pacific Time on April 1, 2025
• Sponsor: Qualcomm Technology Inc.
• Competition: 2025 Low-Power Computer Vision Challenge (2025 LPCVC)
• Vision Task: Image classification for different lighting conditions and styles
• Hardware: The submitted model will be evaluated under the following platforms on Qualcomm AI Hub:

• Snapdragon 8 Elite QRD

• Software: Qualcomm AI Hub
• Technical Support: Subscribe to the newsletter or join the Qualcomm AI Hub Slack workspace. Make sure to join channel #lpcvc for competition related notifications.
• Sample Solution & Referee System: Details see below
• Prizes:
• Champion: $1,500 + a laptop with Snapdragon X Elite processors (Snapdragon X Elite Laptop)
• 2nd: $1,000
• 3rd: $500
• $200 for first 5 teams with valid submissions (better than sample solution)        

2. Prerequisites

Two Registrations:

• Each team is required to register a team account and sign the agreement document (one team only needs to register once). We will use this registration information to manage teams and their submissions.
• Sign up an account on Qualcomm® AI Hub (top right corner). Every team member can register an account if they want. The Qualcomm® AI Hub is a powerful tool for us to compile, profile, and infer images on real mobile devices.

Please refer to our sample solution for track1 for more details of participation.

3. Sample Solution

Participants are required to compile their models on AIHub and share the compiled models with the referee account provided. Once the model is uploaded, participants must complete the submission survey available at [link], providing details such as their model ID, team name, and other relevant information.

• A sample solution and the necessary code for compiling and uploading models can be found on GitHub at https://github.com/lpcvai/25LPCVC_Track1_Classification_Sample_Solution
• A more detailed guide for using AI Hub can be found on GitHub at https://github.com/lpcvai/25LPCVC_AIHub_Guide 

4. Evaluation Details

4.1 Data:

The evaluation data consist of the 64 classes from the 80 classes of COCO image detection. Each class has several images under different lighting conditions. Some classes have images generated with Stable Diffusion.

See the sample data at: https://drive.google.com/drive/folders/1yPhC5FRDYtT9N0U7X015hdbYz30l1mMh?usp=sharing 

4.2 Model Input and Output:

• Model Input: The model input for Track 1 will be RGB images in 224x224 resolution, and the input tensor shape should be (batch, 3, 224, 224). All images should be in [0, 1] in float. The models may include other normalization at the beginning of the model if necessary.
• Model Output: The model output for Track 1 will be an array of logits (same as mobilenetV2, before softmax) of shape (1, 64).

4.3 Metrics:

• Stage 1: Execution Time: For all submitted models, only those with a faster execution time than the base model/sample solution on the targeted test device in Qualcomm AI Hub will be considered a valid solution.
• Stage 2: Accuracy: For valid solutions, we will calculate the prediction accuracy across all test data and rank the results accordingly.
• Final score: Accuracy / Execution Time

See the referee system at:
https://github.com/lpcvai/25LPCVC_Evaluations 

The following is the list of object classes used for the classification challenge. Participants will categorize images into one of these predefined categories based on their content.

Class List:

1. Bicycle
2. Car
3. Motorcycle
4. Airplane
5. Bus
6. Train
7. Truck
8. Boat
9. Traffic Light
10. Stop Sign
11. Parking Meter
12. Bench
13. Bird
14. Cat
15. Dog
16. Horse
17. Sheep
18. Cow
19. Elephant
20. Bear
21. Zebra
22. Backpack
23. Umbrella
24. Handbag
25. Tie
26. Skis
27. Sports Ball
28. Kite
29. Tennis Racket
30. Bottle
31. Wine Glass
32. Cup
33. Knife
34. Spoon
35. Bowl
36. Banana
37. Apple
38. Orange
39. Broccoli
40. Hot Dog
41. Pizza
42. Donut
43. Chair
44. Couch
45. Potted Plant
46. Bed
47. Dining Table
48. Toilet
49. TV
50. Laptop
51. Mouse
52. Remote
53. Keyboard
54. Cell Phone
55. Microwave
56. Oven
57. Toaster
58. Sink
59. Refrigerator
60. Book
61. Clock
62. Vase
63. Teddy Bear
64. Hair Drier