| Title |
Cross-dataset insights for fine-grained vehicle orientation prediction |
| Authors |
Pasaulis, Tomas ; Pečeliūnas, Robertas ; Žuraulis, Vidas ; Raudonis, Vidas ; Sledevič, Tomyslav ; Matuzevičius, Dalius |
| DOI |
10.3390/electronics15102097 |
| Full Text |
|
| Is Part of |
Electronics.. Basel : MDPI. 2026, vol. 15, iss. 10, art. no. 2097, p. 1-27.. ISSN 2079-9292 |
| Keywords [eng] |
vehicle orientation estimation ; fine-grained orientation ; cross-dataset generalization ; domain shift ; label harmonization ; ConvNeXt-Small ; Car Full View dataset ; Freiburg Static Cars 52 dataset |
| Abstract [eng] |
Fine-grained vehicle orientation estimation is widely reported with strong in-domain accuracy, yet performance degrades substantially when models are applied across datasets; the relative contributions of visual domain shift and annotation label incompatibility to this degradation remain poorly understood. A controlled cross-dataset benchmark was conducted using two publicly available datasets—Car Full View (CFV) and Freiburg Static Cars 52 v1.1 (UnsupCar)—under a fixed ConvNeXt-Small predictor with a varied training source, test target, and image preprocessing strategy. All conditions were evaluated with five-fold cross-validation at the vehicle-instance level. Annotation label incompatibility was identified as the dominant source of transfer error: correcting the angular convention mismatch in UnsupCar orientation labels reduced cross-dataset circular mean absolute error (CMAE) by approximately 3.5–4.5∘. Crop protocol was a similarly large factor—train/test crop mismatch raised CMAE into the 9–12∘ range. Square cropping with mirrored boundary padding provided the most robust preprocessing across both in-domain and cross-dataset conditions. After label harmonization, a residual transfer gap of approximately 2∘ remained, with a consistent directional asymmetry favoring the UnsupCar-to-CFV transfer direction. Joint training on both harmonized datasets achieved the best-balanced performance (3.77∘ on CFV; 5.38∘ on UnsupCar). These results demonstrate that instance-level splitting, explicit label harmonization, and consistent crop definition are necessary preconditions for credible cross-dataset vehicle orientation evaluation. |
| Published |
Basel : MDPI |
| Type |
Journal article |
| Language |
English |
| Publication date |
2026 |
| CC license |
|
