ETH Zurich researcher spanning end-to-end driving, incremental detection, and 3D articulation understanding
3
papers here
3
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Not yet scored: this profile's bibliographic coverage is too thin to quantify output. Once aggregated, this axis would measure publication volume and consistency — papers per year and venue types — for the linked ORCID record.
Impact—
No citation data is available yet, so impact is unscored. This axis would gauge scholarly influence through citation counts and uptake of the reviewed papers once citation coverage for this record is sufficient.
Focus0/100
Scores 0, and the raw field distribution explains why: coverage splits evenly at one third each across autonomous-driving robot learning, incremental object detection, and 3D articulation modeling — one reviewed paper per area. Within this three-paper sample the work is deliberately broad, so the zero reflects measured breadth of the sample rather than a verdict on long-term direction.
Collaboration61/100
Scores 61. The reviewed record spans 20 distinct co-authors across three 2026 arXiv preprints, with recurring clusters in ETH Zurich computer vision (Christos Sakaridis, Luc Van Gool, Danda Pani Paudel) and autonomous driving (Kashyap Chitta, Maximilian Igl, Peter Karkus, Boris Ivanovic, Marco Pavone). Schindler sits in the middle author list on all three, none as corresponding author — breadth of ties rather than single-team leadership.
Activity—
Insufficient dated records to score activity. This axis would track publishing cadence — year-over-year rhythm, gaps, and current momentum; the three 2026 preprints alone cannot establish a reliable trend.
Reproducibility55/100
Scores 55, the mean repro-readiness over the three reviewed papers (reproMean 55, reproPapers 3), each rated medium. Evolve's verdicts temper the number constructively: OPTED is called a rigorous, well-controlled study with minor collision-metric blemishes; FAMOS shows large gains but author-adapted baselines; MECAIL matches baselines on D-RICO while lagging on ODinW-13 full training.
Research summary
Konrad Schindler is an ETH Zurich–affiliated researcher whose reviewed output spans three adjacent areas of visual and embodied intelligence: end-to-end autonomous driving, incremental object detection under communication constraints, and feed-forward 3D understanding of articulated objects. Evolve's catalog holds three 2026 arXiv preprints. OPTED introduces on-policy fine-tuning for end-to-end driving with a render-free reinforcement-learning teacher, lifting closed-loop driving scores 1.6× and 9.5× in a study Evolve rates as rigorous and well controlled, with minor collision-metric blemishes. FAMOS tackles feed-forward articulation modeling from sparse multi-state point clouds, reporting large gains whose baselines are, however, author-adapted. MECAIL proposes communication-aware incremental learning with kilobyte-scale (12 KB update; 14.6 KB spatiotemporal experts) model edits, matching baselines on D-RICO while lagging on ODinW-13 full training. Bibliometrically the sampled record is broad but thin: the field distribution splits evenly (one third each) across the three areas, yielding a focus score of 0, while 20 distinct co-authors across the papers give a collaboration score of 61. Reproducibility averages 55 (medium readiness on all three papers). Schindler appears in middle, non-corresponding positions, consistent with a senior collaborator role. Output, impact, and activity axes remain unscored pending more bibliographic coverage.
Collaboration pattern
All three reviewed papers are 2026 arXiv preprints with Schindler in a middle-author, non-corresponding position — a pattern consistent with senior collaboration rather than hands-on lead authorship. The record lists 20 distinct co-authors. Two clusters stand out among frequent collaborators: an ETH Zurich computer-vision group (Christos Sakaridis, Damiano Da Col, Danda Pani Paudel, Luc Van Gool) and an autonomous-driving group (Kashyap Chitta, Maximilian Igl, Peter Karkus, Boris Ivanovic, Marco Pavone), with Matthias Neuwirth-Trapp and Maarten Bieshaar also recurring.
Research keywords
Highlights
Render-free RL teacher lifts closed-loop driving by 1.6×/9.5×
Feed-forward articulation estimation from sparse point clouds
Kilobyte-scale expert updates for communication-constrained detection
Timeline
- 2026MECAIL preprint on arXiv: communication-aware incremental object detection with kilobyte-scale spatiotemporal experts.
- 2026FAMOS preprint on arXiv: feed-forward 3D articulation modeling from sparse multi-state observations.
- 2026OPTED preprint on arXiv: on-policy fine-tuning for end-to-end driving with a render-free teacher, lifting closed-loop scores 1.6×/9.5×.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric scores come from a thin sample: three 2026 arXiv preprints, 20 distinct co-authors, and reproducibility averaged over three papers; output, impact, and activity axes are unscored.
- Profile is name-resolved from public bibliographic data linked to ORCID 0000-0002-3172-9246 and the ETH Zurich affiliation.
- Country-level metadata was not available in the input record and is omitted rather than inferred.
- All cataloged papers are middle-author, non-corresponding arXiv preprints; no peer-reviewed venue or first-author work appears in this sample.